An Evaluation of Rural Electrification Adoption

AN EVALUATION OF RURAL ELECTRIFICATION ADOPTION

DYNAMICS IN MERU-SOUTH SUB-COUNTY, THARAKA-NITHI

COUNTY, KENYA

Mbaka Charity Kageni (B.Ed)

C50/ 20113/2012

A Thesis Submitted in Partial Fulfillment of the Requirement for Award of

the Master of Geography (Urban and Regional Planning) in the School of

Humanities and Social Sciences of Kenyatta University

June, 2015

DECLARATION

This thesis is my original work and has not been presented for a degree or any other award in any university.

Signature………………………….. Date………………2015

Mbaka Charity Kageni (C50/20113/2012)

Department of Geography

Kenyatta University

Supervisors

We confirm that the work reported in this thesis was carried out by the candidate and has been approved for submission with our authority as university supervisors.

Signature…………………………… Date……………….2015

Dr. Philomena Muiruri

Department of Geography

Kenyatta University

Signature……………………………… Date………………….2015

Dr. Kennedy Obiero

Department of Geography

Kenyatta University

DEDICATION

To Dad Mbaka Njeru, Mum Purity Mbaka, my dear sisters Faith and Lillian, and my dear Oscar. I am very grateful for your unending support that has made this degree a reality.

iii

ACKNOWLEDGEMENTS

First, I would like to thank God for his love, mercy, and protection over me throughout this study. Completion of this work has been made possible through the assistance and cooperation of several people. I thank Kenyatta University for the opportunity to do this Masters degree. I humbly extend my sincere gratitude to my supervisors Dr. Philomena Muiruri and Dr. Kennedy Obiero of Kenyatta University for their unfailing academic and professional guidance and support towards the completion of this thesis. I am greatly indebted to my parents and my sisters for their prayers, encouragement and financial support throughout my studies. May I also acknowledge the assistance received from colleagues; Barbara, Jasper, Ann and Mathenge. To my lecturers in the Department of Geography, I am always grateful to you for laying the foundation for my postgraduate Education in

Geography. Finally, I am also grateful to all the field assistants, especially Alfred

Kimathi, and also wish to appreciate the support and responses given by household heads, officials from Kenya Power and Rural Electrification Authority, without whom this work would not have been accomplished. God abundantly bless you all.

TABLE OF CONTENT

DECLARATION...... II

DEDICATION...... III

ACKNOWLEDGEMENTS ...... IV

TABLE OF CONTENTS ...... V

LIST OF TABLES ...... X

ABBREVIATIONS AND ACRONYMS ...... XII

OPERATIONAL DEFINITION OF TERMS AND CONCEPTS ...... XIII

ABSTRACT ...... XV

CHAPTER ONE ...... 1

INTRODUCTION...... 1

1.1 Background to the Study Problem ...... 1

1.2 Statement of the Problem ...... 2

1.3 Research Objectives ...... 3

1.4 Research Questions ...... 4

1.5 Research Hypotheses………………………………………………………...4

1.6 Significance and Justification of the Study ...... 4

1.7 Scope and Limitations of the Study ...... 5

1.8 The Conceptual Framework on Rural Electrification Adoption and Benefits .. 6

CHAPTER TWO ...... 8

LITERATURE REVIEW ...... 8

2.1 Introduction ...... 8

2.2 Overview of Rural Electrification Programmes ...... 8

2.3 Household Socio-economic Factors Influencing Electricity Connection ..... 10

2.4 Socio-economic Benefits of Electricity to Households ...... 14

2.5 Uses of Electricity among Households ...... 16

2.6 Effect of Rural Electrification on Public Facilities ...... 17

2.7 Spatial Distribution of Electricity Accessibility and Adoption ...... 17

CHAPTER THREE ...... 20

RESEARCH METHODOLOGY ...... 20

3.1 Study Area ...... 20

3.2 Research Design...... 22

3.3 Variables of Study...... 22

3.4 Pilot Study ...... 22

3.5 Research Assistant Selection and Training ...... 23

3.6 Sampling Method and Procedure ...... 23

3.8 Data Processing and Analysis ...... 26

3.8.1 Descriptive Statistics Analysis ...... 27

3.8.2 Non-parametric Test ...... 27

3.9 Ethical Considerations ...... 30

3.10 Interpretation and Presentation of Results ...... 31

CHAPTER FOUR ...... 32

RESULTS AND DISCUSSISON ...... 32

4.0 Introduction ...... 32

4.1 Socio-economic Characteristics of the Respondents ...... 32

4.2. Household Socio-economic Characteristics Influencing Electricity Adoption in Meru-South Sub-County………………………………………………………34

4.2.1 Status of Electricity Adoption ...... 34

4.2.2 Gender Characteristics of Adopter and Non-Adopter Household Heads

……………………………………………………………………...... 35

4.2.3 Age Characteristics of Adopters and Non-Adopters Household Heads .

……………………………………………………………………………….35

4.2.4 Educational Characteristics of Adopters and Non- Adopters ...... 36

4.2.5 Occupational Characteristics of the Household Heads ...... 38

vii

4.2.6 Household Size of Adopters and Non-Adopters ...... 40

4.2.7 Marital Status of Adopters and Non-Adopters‘ Household Heads ..... 40

4.2.8 Characteristics of Household‘s Dwelling House by Type of Wall

Material ...... …………….41

4.2.9 Main Sources of Income of the Household Heads ...... 42

4.2.10 Monthly Total Income of Adopters and Non-Adopters ...... 44

4.3 Extent to which Households‘ Socio-Economic Characteristics Influence

Electricity Adoption………………………………………………………………45

4.3.1 Distance from the Transformer ...... 47

4.3.2 Education attainment ...... 48

4.3.3 Income...... 49

4.3.4 Household Size ...... 49

4.3.5 Gender ...... 50

4.4 Socio-economic Benefits and Challenge, of Rural Electrification Adoption

among Households ...... 51

4.4.1 Uses of Electricity among Households ...... 53

4.4.2 Electricity Benefits from Home Business ...... 58

4.4.3 Ranking of the welfare benefits of electricity ...... 61

4.4.4 Challenges in Electricity Adoption in Meru-South Sub-County ...... 62

4.5 Effect of Rural Electrification on Public Facilities ...... 67

4.5.1 Quality of Service Provision in Electrified and Un-electrified Public

Facilities ...... 69

viii

4.6 Spatial Distribution and Characterization of Transformers, Adopters and Non

-adopters and in Meru-South Sub-County ...... 74

4.6.1 Spatial Distribution and Characterization of Transformers in Meru- South

Sub-County ...... ……….74

4.6.2 Spatial Distribution of Adopters in Meru-South Sub-County ...... 78

CHAPTER FIVE ...... 82

SUMMARY OF FINDINGS, CONCLUSIONS AND RECOMMENDATIONS ...... 82

5.1 Introduction ...... 82

5.2 Summary of Key Findings ...... 82

5.3 Conclusion ...... 87

5.4 Recommendations ...... 89

5.5 Areas of Further Research ...... 90

REFERENCES ...... 91

APPENDICES ...... 98

APPENDIX I: SAMPLE QUESTIONNAIRE ...... 98

APPENDIX II: IN-DEPTH INTERVIEW GUIDE INVOLVING ELECTRICITY AGENCIES /LOCAL ADMINISTRATORS IN MERU - SOUTH SUB-COUNTY ...... 103

APPENDIX III: TRANSFORMER ATTRIBUTE SHEET ...... 104

LIST OF TABLES

Table 3.1: Variables for Data Collection ...... 30

Table 4.1: Summary of Socio-economic Characteristics of the Household Heads33

Table 4.2: Gender Characteristics of Electricity Adopters and Non-Adopters .... 35

Table 4.3: Link between Education and Age Status of Adopters and Non-Adopters ...... 38

Table 4.4: Household Size Characteristics of Adopters and Non-Adopters ...... 40

Table 4.5: Monthly Total Income of Adopters and Non- Adopters ...... 44

Table 4.6: Test for Multicollinearity ...... 45

Table 4.7: Logistic Regression Analysis of Explanatory Variables in Electricity Adoption ...... 46

Table 4.8: Number of Years Households Connected to Grid Electricity ...... 51

Table 4.9: Electric Appliance Ownership among Electricity Adopters ...... 53

Table 4.10: Electricity Benefits from Appliances in the Households ...... 54

Table 4.11: Frequency Use of Electric Appliances ...... 56

Table 4.12: Importance of Electricity to Children‘s Education ...... 62

Table 4.13: Household‘s Perceptions on Challenges of Electricity Adoption...... 63

Table 4.14: Alternative Sources of Energy in the Households ...... 66

Table 4.15: Responses on Electrified and Un-Electrified Public Facilities in Meru- South Sub-County………………………………………………………...66

Table 4.16: The Mean Scores and Standard Deviation of Quality of Service Provision in Electrified and Un-electrified Public Facilities in Meru-south Sub- County..……………………………………………………………………..68

Table 4.17: Independent Samples t-test for the Differences between Electrified and Un- Electrified Public Facilities…………………………………………………69

LIST OF FIGURES

Figure 1.1: Conceptual Framework on Rural Electrification Adoption and Benefits …………………………………………………………………………………...... 6

Figure 3.1: Map of Meru-South Sub-County ...... 20

Figure 3.2: Sampling Strategy of Households in the Meru South Sub-County .... 25

Figure 4.1: Age Distribution of Adopters And Non-Adopters ...... 36

Figure 4.2: Educational Levels among Electricity Adopters and Non- Adopters …………………………………………………………………………………….37

Figure 4.3: Occupational Characteristics of Household Heads ...... 38

Figure 4.4: Marital Status of Adopters and Non-Adopters ...... 41

Figure 4.5: Household Distribution by Wall Type ...... 42

Figure 4.6: Sources of Income among Adopters and Non-Adopters ...... 43

Figure 4.7: Households Connected to Electricity in Year 2009-2012 ...... 52

Figure 4.8: Distribution of Small Business Activities in Households ...... 59

Figure 4.9: Average Income Generated from the Small Business ...... 60

Figure 4.10: Uses of Income Generated from Small Business ...... 61

Figure 4.11: Installation of Transformers over Years ...... 75

Figure 4.12: Spatial Distribution of Transformers in Meru-South Sub-County ... 76

Figure 4.13: Spatial Distribution of Adopters in Meru-South Sub-County ...... 79

Figure 4.14: Spatial Distribution of Non-Adopters in Meru-South Sub-County . 80

ABBREVIATIONS AND ACRONYMS

GIS: Geographical Information Systems

GoK: Government of Kenya

GPS: Global Positioning System

IEA: International Energy Agency

KP: Kenya Power

KNBS Kenya National Bureau of Statistics

HH Household Head

MSDDP: Meru-South District Development Plan

RE: Rural Electrification

REA: Rural Electrification Authority

REP: Rural Electrification Programme

SPSS: Statistical Package for Social Sciences

SSA: Sub-Saharan Africa

VIF: Variation Inflation Factor

WEO: World Energy Outlook

xii

OPERATIONAL DEFINITION OF TERMS AND CONCEPTS

Rural electrification: Refers to provision and distribution of electricity in rural areas via the national grid extension.

Adoption: Taking up electricity through connection via the national grid extension, and continued use of electricity by households.

Non-adoption: A household without electricity connection from the grid extension.

Electricity accessibility: Households within 600 meter radius of power distribution points (existing functional transformers).

Household: Refers to a group of persons who reside in the same homestead/compound but not necessarily in the same dwelling unit and are managed by the same household head.

Socio-economic characteristics: Refers to demographic characteristic of a household expressed in statistical form such as age, gender, educational status, income level, marital status, and occupation.

Household head: This is the most responsible/respected member of the household who makes key decisions in household on a day to day basis and whose authority is honored by all members of the household

xiii

Rural Development: Sustainable socio-economic improvement of the rural population at household and community level due to among others the productive use of electricity.

Electricity Access: This means that following aspects are in place: distribution network within vicinity, connection to the location of household and supply is operational.

Spatial Distribution: Actual location and arrangement of phenomena on the earth‘s surface

Walling Material: Refers to the building material used to construct household‘s main dwelling unit wall.

xiv

ABSTRACT

Electricity services are crucial for human well-being and to a country‘s socio- economic development. Despite its importance, low levels of electricity adoption and use continue to prevail in most rural areas in SSA. Low socio-economic development has been attributed among others factors, to lack of modern energy sources especially electricity among households, which has been identified as a major setback in propelling empowerment and development at household and community level. There is minimal or no research conducted to understand the socio-economic dynamics of electricity adoption among households in Meru-South Sub-County. The objective of this study was therefore to analyse and identify determinants of electricity adoption, assess the socio-economic benefits and challenges of electricity adoption, assess the effect of rural electrification on development of public facilities and examine spatial distribution of electricity adopters, non-adopters and transformers in Meru-South Sub-County. To achieve these, household interviews were conducted from 150 randomly selected households using closed and opened ended questionnaire. In-depth interview guide was used to collect information from two Rural Electrification officials, two from Kenya Power and two local administration officers. A GPS set was used to geo- code adopter, non-adopter and transformer points. Data collected was statistically analyzed using descriptive statistics. Chi-square and t-test were used to test the magnitude of the association between dependent and independent variables. Logistic regression was used to predict the socio-economic factors influencing electricity adoption in households using statistical package for social sciences (SPSS) programme version 19.0. Data from GPS sets was organized into a compatible file and imported into ArCGIS 10.2 to generate maps. Results showed that 36% and 64% of the respondents in the study area were electricity adopters and non-adopters respectively. Possible predictor factors that significantly influenced adoption were found to be distance from the transformer (p=0.000), education status (p=0.020), gender (p=0.045), household size (p=0.009), and income (p=0.011). Besides low electricity adoption, electricity benefits and potentials among the adopters including improved quality of life through lighting (100%) and businesses (38.8%) among others were revealed. Results revealed a significant difference (p<0.05) in quality of service provision in electrified and non-electrified schools, hospitals, market centers and factories. Results indicated the greatest prior challenges to electricity connection were accessibility (proximity of the transformer) and cost of connection. The transformers were revealed to be in the upper and middle areas compared to lower areas. Adopters were mainly in upper zones while non-adopters were distributed in lower and upper zones. The findings indicate that electricity was not extensively used for income generating services. These results indicate need for promotion of productive uses of electricity in Meru- South which can increase the productivity at household and community levels. This study provides a guide during planning for rural electrification in order to increase electricity adoption and enhancing use for income generating activities.

CHAPTER ONE

INTRODUCTION

1.1 Background to the Study Problem

Electrical energy is a critical facet to a country‘s socio-economic development as well as human socio-economic well-being. However, the International Energy Agency

(IEA, 2012) reported that nearly 1.3 billion people have no access to electricity globally, 85% of this living in rural areas. Of this population, 95% are either in Sub-

Saharan Africa or Asia. There are large variations in electrification rates across and within regions. Transition economies and countries belonging to the Organization for

Economic Co-operation and Development (OECD) have virtually universal connectivity rate (IEA, 2010). North Africa connectivity level stands at 99%, Latin

America 93%, East Asia and the Pacific 90%, the Middle East 89%, while South Asia and Sub-Saharan Africa electrification level is at 60% and 29% respectively. The populations without electricity in these two regions account for 83% of the total world population without electricity (IEA, 2010).

Sub-Saharan Africa (SSA) stands as the least electrified region globally, with close to

585 million of its population having no access to electricity. Generally, the overall electrification level in SSA region stands at 30.5% (59.9% urban; 14.3% rural) with the bulk of un-electrified areas massively stretching into rural areas (IEA, 2010).

Further, it is projected that the population without electricity connection in SSA will increase by 11% to 655 million by the year 2030 (IEA, 2012). It is anticipated that the worsening trend will persist unless robust actions are taken to expand growth in the power sector (IEA, 2012).

Like most other SSA countries, Kenya‘s electricity accessibility and eventual adoption is quite low compared to other countries in the world (World Bank, 2010).

Apparently, countrywide electricity access has slightly improved over years but adoption rate still low (19.2%) which is much lower than the average adoption in SSA at 30.5% (GoK, 2014). Conversely, disparities in the electricity distribution in Kenya are quite high. For instance, electricity connectivity in urban areas (51.3%) is high when compared to rural connectivity (5%) (IEA, 2010). During the 2009 National

Census in Kenya, Meru-South Sub-County was ranked amongst the least electrified region in the country with only 4.8% of rural households in the area having electricity connection (KNBS, 2009). During this same year, REA initiated programmes to accelerate rural electrification via grid extension in the area. Currently, electricity accessibility in the area is reported to have increased over the last five years (REA,

2013). Despite the efforts of electrifying the region, low levels of electricity adoption among households still prevail in the region (REA, 2013). Owing to electricity‘s critical role in rural and household socio-economic development, understanding factors that translate to this low adoption coupled with connectivity patterns in the region is quite essential.

1.2 Statement of the Problem

Socio-economic development in Meru-South Sub-County continues to lag behind despite the area‘s high agricultural and industrial development potential. Among many other bio-physical factors, low development in the region has been attributed to insufficient use of modern energy sources occasioning continued underproduction and decline in economic production. Electricity in particular, has been identified as a prime determinant of household socio-economic development. However, electricity‘s

low accessibility coupled with low adoption present a major setback in propelling and

empowering rural household development in Meru-South sub-county. Inception and

subsequent implementation of REP in 2009 has foreseen substantial increase in

electricity connectivity and accessibility. However, electricity adoption among

households is quite low and below the targeted adoption level for the first phase of

rural electrification. This has been a major drawback since the anticipated and

projected goals of increasing electricity accessibility in the area have not been

realized. Additionally, there is limited focus and lack of detailed information on

adoption dynamics of rural electrification via national grid extension among

households in Meru-South Sub-County.

Based on this background, this study sought to examine determinants of electricity

adoption, examine spatial dynamics of electricity access points, adopters/non-

adopters, as well as the effects of electricity on households and public facilities in the

study area.

1.3 Research Objectives

The broad objective of this study was to evaluate rural electrification adoption

dynamics in the study area. To achieve this objective the study addressed the

following specific objectives:

i. To determine the influence of household socio-economic characteristics on

electricity adoption. ii. To assess the socio-economic benefits and challenges of rural electrification

adoption among households. iii. To assess the effect of rural electrification on development of public facilities.

iv. To examine the spatial distribution and characterization of electricity

adoption, non- adoption and accessibility in Meru-South Sub County.

1.4 Research Questions

This study sought to specifically answer the following research questions:

i. How do households‘ socio-economic characteristics influence electricity

adoption in the study area? ii. What are the socio-economic benefits and challenges of rural electrification

adoption among households in the study area? iii. To what extent is the effect of rural electrification on development of public

facilities in the study area? iv. What is the spatial distribution of electricity adoption, non-adoption and

accessibility, among households in Meru-South Sub-County?

1.5 Research Hypotheses

The study was guided by the following hypotheses:

Ho1: There is no significant relationship between households‘ socio-economic

characteristics and electricity adoption in the study area.

Ho2: There is a no significant difference in quality of service provision in electrified

and un- electrified public facilities in the study area.

1.6 Significance and Justification of the Study

Rural electrification enhances quality of life among households and broadly

stimulates socio-economic development. For instance, electricity provides the basis

for improving productivity by facilitating income generating activities and provides indirect social benefits such as greater equity and improved quality of life.

Understanding electrification networks, connectivity adoption and non-adoption dynamics form a critical aspect in optimal planning of rural development. This would further enhance the understanding of the causes of low adoption despite increased electrification as in the case of Meru-South Sub-County. The study evaluated the socio-economic determinants of rural electricity adoption and household participation in order to unravel the causes of the low share of electrified households in the study area. It is envisaged that the study findings and recommendations will be a benchmark of developing evidence based policies as well as strengthen institutional frameworks that would viably accelerate grid-based electricity adoption in the study area. This will further contribute in the realization of Millennium Development Goals and vision

2030 whose critical driver is electrical energy.

1.7 Scope and Limitations of the Study

The study was conducted in Meru-South Sub-County in Tharaka Nithi County. The focus of this study was on rural electrification via the national grid extension and not other power sources. The limitation the researcher encountered in the study was: the lack of readily available spatial data set on households‘ connection status to the grid and electricity distribution network. There was also a general skeptical view of the 5 respondent households toward a research seeking to probe their monthly income and the household size in relation to the socio-economic characteristics. However, efforts were put in place to mitigate their impacts on the results of the study.

1.8 The Conceptual Framework on Rural Electrification Adoption and

Benefits

The rural electrification programmes trigger unprecedented increase in electricity accessibility across rural areas (Ying, 2006). Enrique (2010) suggests that grid extension is often the cheapest way to reach new consumers and increase access rates among rural communities. However where electricity becomes available, the up take is variable. Kowsari, (2011) postulated that household electrification depends on household characteristics. When electricity is available, it takes between 1 and 3 years for households to get connected. But there are still high percentages that do not connect electricity to their households even after this period (Bhattacharyya, 2013).

Adoption and continuous use of electricity by households enhances quality of life at rural household level and stimulates economy at community level (Khandker, 2009)

(Figure 1.1).

Figure 1.1: Conceptual Framework on Rural Electrification Adoption and Benefits

Rural Electrification accessibility

HOUSEHOLD Enhanced Socio- CHARACTERISTICS Electricity economic benefits and · Income Adoption Empowerment · Occupation · Education level · Gender · Household Size · Age Spillover Benefits · Marital status Electricity Non- · Proximity of Adoption Low Socio-economic Household from Endowment/ transformer Empowerment · Wall material I np ut Increased Adoption Socio-Economic Empowerment Output & Development Better policy making

Source: Adopted and modified from Kowsari, (2011).

Changes at the household and community levels interact and reinforce each other, finally becoming a self-sustaining process. Gaining a better understanding of household electricity adoption determinants enhances understanding in making a distinction of policy that ought to be used to improve adoption where electricity is accessible. The implementations of specific policies contribute to rural development and social economic empowerment among rural households. These links are illustrated in the above conceptual framework. Consequently this study adopts this conceptual framework by examining households‘ characteristics influencing electricity adoption, and socio-economic benefits of rural electrification.

CHAPTER TWO

LITERATURE REVIEW

2.1 Introduction

This chapter reviews key findings, theoretical and methodological contributions by past studies in the subject matter being studied. The thematic areas reviewed include; synopsis of various rural electrification programs and success accounts, determinants of rural electrical energy adoption and its socio-economic benefits among households.

This review also links REP to the overall rural development and identifies gaps which inform this study.

2.2 Overview of Rural Electrification Programmes

Electrical energy is one of the prime inputs for social and economic development globally (Katuwal et al., 2009) and remains partly a fundamental prerequisite to economic development (Ahlborg et al., 2011). Most developing countries underpin the need for RE policies in intensifying RE programs, especially in Sub-Saharan

Africa where over 585 million people lack access to electricity (IEA, 2011).

Electricity access is increasingly at the forefront of governments‘ preoccupations, especially in the poorest countries. As a consequence, a lot of rural electrification programs and national electrification agencies have been created in these countries to monitor more accurately the needs and the status of rural development and electrification. Nonetheless, success stories of rural electricity have been reported world over. For instance, China implemented the RE program and currently has only

0.6% population without connectivity. Through this program, China adopted a six- phase framework aimed at developing locally managed electrical programs, and combination of central grid extension, local grids and off-grid solutions (Zhang et al.,

2011). Conversely, cooperatives and government electricity providers in Costa Rica expanded RE increasing its adoption to almost 100% (Barnes, 2007). Bangladesh has experienced a more balanced approach towards rural electrification and subsequent success by underpinning to power residential units and advocate for optimal productive usage (Barkat, 2003). Brazil has made significant advancements by enhancing RE accessibility that the country currently nears 100% of RE (Niez, 2010).

The Brazilian RE success is principally attributed to effective regulations that integrate provision of affordable electricity to low-income consumers with the aim of promoting rural development and satisfying social demands.

North Africa is the region with highest connectivity (99%) in Africa. Tunisia‘s is the most electrified country (99.5% connectivity) in this region (IEA, 2010). Tunisia success has been due to state commitment, integrated rural development process, effective institutional approach and effective tariff policy (Cecelski et al., 2006). SSA has experienced low rural electrification rates (only 14.3% connectivity). Despite significantly low rates of RE accessibility in Africa, some countries have substantially increased RE among them, for instance Ghana (with urban and rural accessibility at

99% and 49% respectively). The Ghanaian success story has been attributed to long- term energy planning with clear targets, availability of external funding, and active central government participation in the implementation of energy policies

(Kemausuor et al., 2012). In South Africa, technology development through prepaid metering and blanket electrification, played an essential role in reducing the real cost per connection and thus contributed to the attainment of social objectives of the electrification program (Bekker et al., 2008). Evidently, successful RE programmes have been realized under intense political support, participation of special institutions and local committees.

In Kenya the rural electrification programme was started in 1973. However, despite these efforts, many parts of the country have remained unconnected to electricity. By

2003, that is 30 years since the inception of the programme, overall connectivity rate in rural areas was standing at 4%. This realization led to more allocation of funds for the programme which in turn led to the established of the Rural Electrification

Authority (REA) under section 66 of Energy Act 2006 (No. 12 of 2006). REA is a single autonomous authority with a clear mandate to accelerate the pace of electrification in rural areas (REA, 2013). The government‘s intentions are that all

Kenyans have access to electricity by the year 2030 as envisaged in the country‘s long term blueprint for development, the Vision 2030. Two main methods adopted for RE program in Kenya are grid extensions and promotion of renewable energy sources.

REA works in partnership with Kenya Power in maximizing household‘s connection in areas supplied with electricity, REA‘s connection rate now stands at 300,000 persons per annum. From the year 2007 to 2013 (during REA‘s operations), 23167 out of 25837 prime facilities identified in the master plan were connected to electricity translating to about 90% connectivity (REA, 2013). Evidently many households have not been connected to electricity hence the purpose of the study.

2.3 Household Socio-economic Factors Influencing Electricity Connection

Electrical energy for household use remains an important avenue of alleviating rural poverty especially in Developing Countries (Abdullah et al., 2012). There is continued emphasis on increasing rural electricity accessibility to rural households globally whose adoption at household level is quite low in the SSA (Kemmler, 2007).

However, increased accessibility to RE does not translate into automatic adoption by rural households (Christensen et al., 2012). Additionally, accessibility to a grid connection does not guarantee use of electricity among all end users (Winkler, 2011).

For instance, low connection rates within grid-electrified villages of 12%, 39% and

30% in Botswana, Ethiopia and Senegal respectively, have been documented

(Ketlogetswe et al 2007; ESMAP, 2007; Bernard et al., 2009). Assorted factors account for the decision to adopt or not to adopt RE among households. Although macro-level factors influence adoption, micro-trends cannot be accurately extrapolated from national figures; especially among poor rural households, where adoption patterns may easily be obscured by other factors (Elias et al., 2005). This study will explore the determinants of electricity adoption at the household level in order to determine micro level determinants of adoption among the rural households.

There is unanimous consensus that households‘ head income is a major factor that determines adoption of electricity in the residential sector (Abdullah et al., 2012;

Barnes et al., 2005; Mishra, 2010). Studies further show income to be a prime driver of electricity adoption; reporting strong correlation between income increase and adoption of electricity (Barnes, et al., 2005; Johansson et al., 2004; Pachauri, 2007).

According to Mishra (2010), income cannot be a key determinant of electricity adoption and also there is a negative correlation between electricity adoption and income. For instance, Ketlogetswe et al. (2007), indicated that connection rates in

Botswana were very low despite friendly payment systems; 10% covered at installation and 90% of the remaining cost spread across 10 years. Even with subsidies, connection rates remained low, suggesting that price policies may not fully explain the observed low levels of electricity adoption. Additionally, loan acquisition

to facilitate electricity adoption is quite limited, particularly among the poor households, who are reluctant to engage in long-term financial commitments

(Bernard, 2012).

Both endogenous and exogenous factors determine adoption of RE. Endogenous factors are sub- categorized as economic and non-economic characteristics which reflect the capabilities, behavioral, cultural characteristics, attitudes, preferences and experiences of households (Kowsari, 2011). Exogenous factors are external conditions that influence household decisions regarding their choice of energy systems and associated incentives. Such factors include physical environment, policies and regulations (Barnes et al., 2005; Victor et al., 2009). Energy supply factors such as, availability, accessibility and reliability also influence choice and adoption of electricity (Barnes et al., 2005). Availability of power grids not only affect electricity adoption but also change service demands among households (Davis

1998; Heltberg, 2005).

Information, education, and social learning are also described as factors that determine rates of adoption electrical energy systems. Lack of information regarding the alternative energy systems and the associated benefits have been shown as a barrier towards adoption (Heltberg 2003; Schlag et al., 2008; and Whitfield, 2006).

Research has explained that outsourcing, social economic factors, physical infrastructure and financing are likely to influence connectivity to the national grid

(Wanyoike, 2012). Rural households‘ perception of the benefits of electricity is also a factor that determines electricity adoption. Although most studies find an important demand for electricity, households mainly perceive it as a luxury service rather than a

so-called productive investment (Peters et al., 2009). In addition fears of vaguely understood billing system, and delayed installation and erratic supplies also undermine connectivity (Mbatia 2011; Bernard, 2012).

Gender roles substantially influence decision making on energy at the household level. According to Clancy (1999), though women are the main end-users of energy they are limited in their involvement in planning and implementation levels of most of the projects in the energy sector. Often women are not in a position to make or influence decisions concerning energy use (Clancy, 1999).

A female head of the household might be the main and only provider, not only for the needs of their dependents, but also for their own needs. In other words, female household heads tend to have more responsibilities than other females who normally have support from their spouses or partners (Moghadam, 1998). In addition, female household heads tend to face labour market disadvantages and time constraints because of tasks relating to the upkeep of the household and this makes it difficult for them to earn sufficient income (Fuwa, 2000). Nishimwe et al. (2014) found out that female-headed households were consistently less likely to be connected to the main source of electricity than those headed by males.

Distance of the household from the transformer is a determinant of electricity adoption among households (Andreas, 2006). Transformers act as electricity access points for rural households (REA, 2013). Distance of the households to a nearest transformer is measured to determine the upfront cost of connection. The closer the transformer is to a dwelling, the more likely the households will be connected to the

power grid. Standards set by the energy utility indicate that households within 0.6 km from a transformer get a subsidized cost of connection to the power grid.

Conclusively, a pragmatic approach of determining RE, its benefits and increased adoption is essential thus required in this study. As such this study endeavored to critically assess the determinants of electricity adoption at the household level in the study area and report on the diverse issues that influence adoption and non-adoption, and identified optimal measures needed in enhancing rural electrical adoption at the micro level. In essence incorporating a range of endogenous and exogenous characteristics avoids overemphasizing on selected determinants over other possible factors.

2.4 Socio-economic Benefits of Electricity to Households

Gaining a better understanding of rural electrification‘s benefits helps clarify the framing of policies and options for developing countries (Barrios, 2008). It is universally accepted that electrification enhances quality of life at the household level and stimulates economy at a broader level (Khandker et al., 2009). Electricity is a critical tool for use at micro level (household). Studies have revealed that energy services are a crucial input to the primary challenge of providing basic needs among households. For instance, household access to electricity facilitates timely cooking of food, provides a comfortable living temperature, lighting, and enables the use of communication appliances, which all contribute to the individual and family quality of life (Mvondo, 2010).

Numerous uses of electricity by households that are seen as consumptive (e.g., home lighting or television) in fact help achieve political freedom, opportunities to receive basic education, opportunities to receive health care, and freedom to participate in the labor market (Cabraal, et al.,2005). Studies carried out in Philippines, and India on grid-based electrification showed that increased income-generating activities are positively correlated with electrification; also that electrification increases chances of households engaging in productive use (Fishbein et al., 2003). In rural areas, electricity could be used for crop irrigation, agro-processing, small-scale mining and to facilitate tourism. According to Peters et al (2009) households are frequently not aware of the economic potentials of electricity and hence cannot be expected to consider the grid connection decision and the usage of electricity rationally in an economic sense. Gaining a better understanding of rural electrification‘s benefits helps clarify the framing of policies and options for Developing Countries (Barnes et al., 2005).

Most studies that have been carried out on benefits of rural electrification have been on impact assessment. Few studies have evaluated the direct appliance driven socio- economic benefits among rural households. In addition most studies carried out on benefits of electrification have been too general mostly focusing on the national level more so data is usually collected from urban areas, hence the focus on local levels and especially in rural areas have been inadequate. This study seeks to study the benefits at local level further guided by the principle that assert conducting a survey at periodic intervals to track the progress of a rural electricity project over time, would wholly or partially be replicated in later years to yield time series data on the impact of the rural electrification project in the study area (ESMAP, 2007).

2.5 Uses of Electricity among Households

Rural development is referred to as development that benefits rural populations; where development is understood as the sustained improvement of the population‘s standards of living or welfare (Anríquez et al., 2007). Rural electrification programmes that underpin development of both residential and productive uses have recorded significant success (Barkat, 2003). Cabraal et al. (2005) noted that access to electricity has a significant impact on rural development only when it is used efficiently and on income-generating activities. Access to electricity has a substantial positive impact on rural growth and livelihoods. In terms of economic development, it provides the basis for improving productivity by facilitating income generating activities and improving the business climate. Rural areas have three categories of energy use: household energy, agricultural energy, and energy for small/micro enterprises (Karekeziet al., 2003).

Productive use of energy in rural areas is expected to result in increased rural productivity, greater economic growth, and a rise in rural employment, which would not only raise incomes but also reduce the migration of the rural poor to urban areas

(Cabraal et al., 2005). With respect to agricultural production, electricity use result to modernized agriculture providing motive power for agriculture-based industries and power farm machinery (Cabraal et al., 2005).

Electricity provides significant indirect social benefits such as greater equity and improved quality of life. A study in Bangladesh revealed women in households with electricity were much more aware about gender equality issues than women in households without electricity (Barkat, 2003). Electrification programs should

therefore be combined with a strong development focus and target productive sectors

(Christensen et al., 2012). This study endeavored to find out how electrification substantially affects rural households.

2.6 Effect of Rural Electrification on Public Facilities

Access to electricity in public facilities is critical to service delivery. Most rural electrification programmes during the initial implementation mainly target public facilities in rural areas. Public facilities in rural areas are mainly found in rural centres, often called ‗growth points‘. In these centers we find government infrastructure such as police stations/posts, agricultural extension, and public facilities such as schools, trading centres, health centres, administrative centers, churches mosques, boreholes and other public facilities. These are known as centres of rural development. Supplying electricity to regions with public facilities enhances the effect of existing public facilities and improves the overall utility of these centers to surrounding areas (Barnes, 2007).

2.7 Spatial Distribution of Electricity Accessibility and Adoption

Spatial distribution is the arrangement of phenomena across earth‘s surface. Mapping allows discovering of new relationships and patterns for geographically referenced information (Wang and Luo, 2005). Observations are made to describe the geographic patterns of features, both physical and human across the earth. The three general categories of geographical patterns conventionally used as benchmark to describe how points structure spatially include; cluster (aggregate), disperse and random patterns.

One of the ways of capturing spatial information is through combination of GPS data collection and mapping software. This technique enables researchers to produce more than just location of features, but also discover new patterns and relationships for geographically referenced data points. For instance, a study carried out in India to investigate the causes of the spatial disparities in electrification rates, mapping based on the census data portrayed a remarkable spatial heterogeneity in electrification rates, some areas having a higher share of households without access to electricity than others. A study by Ogalo (2011) revealed fair connection levels among households in

Nyamarambe Division; however the study did not establish the distribution patterns of adopters and non-adopters across the region. A study of Machakos County revealed that 93% of electricity adopters were within a radius of two kilometers from the main electricity grid as well as the tarmac roads (Kembo, 2013). This study depended wholly on the self reported information without any spatial analysis which would otherwise give a visual presentation for further interpretation. Mapping based on GIS shows patterns in an instant that might not be apparent from a table of figures holding the same information. For that reason, GIS mapping is a powerful method of presentation, especially for policy purposes (Sieber, 2006). According to Gibson and

McKenzie (2007) collection of GPS coordinates should become a routine part of household survey collection, since doing so can lead to better economics and better policy advice. Consequently, this allows visually isolation and examination of spatial distribution patterns in the study area.

With vast majority of world‘s population lacking access to electricity assumed to be concentrated in rural areas the distribution patterns remain unknown in many developing countries (IEA, 2012). Unfortunately most of household survey studies have not incorporated geo-spatial techniques in understanding the spatial distribution

(Gibson 2007). Therefore this study seeks to map and examine the spatial distribution of adopters and non-adopters‘ households and the electricity accessibility points in the study area. GIS is eminently useful as a data visualization tool, but its capabilities also make it useful in data analysis. Evidently GIS is useful in the analysis of statistical differences among regions and in the analysis of spatial patterns of respondents.

CHAPTER THREE

RESEARCH METHODOLOGY

3.1 Study Area

The study was carried out in Meru South Sub County, of Tharaka - Nithi County

(Figure 3.1).

Figure 3.1: Map of Meru-South Sub-County

Source: Modified from MSDDP (2007)

The area lies between longitudes 37 18‘37‖and 37 28‘33‖East and Latitude 00 07‘23‖ and 00 26‘19‖South. Meru-South lies in the Upper zones-LH1, UM1, UM2, Middle zones-UM3 and Lower zones-LM3, LM4, LM5 (Jaetzold et al., 2006) on the eastern slopes of Mount Kenya. The altitude ranges from 830 metres in the lower areas to

1850 meters above sea level at the base of Mt. Kenya. Annual mean temperature range between 14Co to 17Co in the upper and middle areas while, in lowlands it ranges between 22oC to 27oC .The rainfall is bimodal with long rains (LR) occurs from

March to June and short rains (SR) from October to December. The upper areas experience reliable rainfall while middle areas, medium rainfall and lower regions unreliable and poorly distributed rainfall.

Meru-South has a population density of 205 persons per Km2. The population density varies within the Sub-County with Chuka Division at 316 persons per Km2 followed by Magumoni Division with a population density of 189 per Km2 and Igamba

Ng‘ombe with the least population density of 110 persons per Km2. The topography of the Sub-County is influenced by the volcanic activity of Mt. Kenya. Numerous rivers which originate from Mt. Kenya Forest traverse the Sub-County and flow eastwards as tributaries of Tana River, which discharge its water into the Indian Ocean.

The major economic activities, the livelihood systems engaged by the local community include; agriculture and livestock production. Coffee and tea are major cash crops, while maize, beans, potatoes, cassava and bananas are grown for subsistence and to some extent cash sale. Livestock keeping is also practiced where households keep dairy cattle goats and sheep and poultry (Development plan, 2006).

Public infrastructures are moderate unevenly distributed in the area. The prime public infrastructure and facilities include: primary and secondary schools, hospitals, markets, primary health centers, roads, electricity, and agro-service centers.

3.2 Research Design

The study utilized a survey design approach in combination with mixed methodologies combining both qualitative and quantitative research methods to enable an in-depth investigation into the subject matter studied. Additionally, GPS was used to record where transformers had been erected and the interviewed household points.

3.3 Variables of Study

A household was considered either as an electricity adopter or non-adopter. Selection of independent and dependent was based on an assumption that they were relevant in influencing adoption potential of households in the study area. During this study the dependent variable was electricity adoption status, whereas the independent variables were: household socio-economic characteristics such as age, gender, occupation, educational status, and household size, marital status, house type, sources of income, monthly total income, socio- economic benefits, and also perception on quality of service provision.

3.4 Pilot Study

A pilot study was undertaken one month prior to the actual study to test the validity, reliability and sustainability of the research instruments. A sample of 5 household from Chuka, Magumoni and Igamba Ng‘ombe division were randomly selected and interviewed. Two officers from Kenya Power and Rural Electrification Authority participated during the pre-testing exercise. The findings of the pilot study helped to make adjustments on the questionnaires. Clarification of some questions that were not clear was done to ensure that they facilitated collection of adequate and reliable data.

3.5 Research Assistant Selection and Training

The study employed one research assistant based on academic qualification, field work experience and who had ability to communicate in the local language.

Consequently, a one day training was given on the significance of each question in the questionnaire as far as the research was concerned, how best to introduce the topic to the respondent and collect accurate information from the respondents. More so, the research assistant was informed on research ethical consideration such as respondents‘ anonymity among others. The training was concluded with my demonstration of how to administer the questionnaire in the field to the respondents.

3.6 Sampling Method and Procedure

Meru-South was selected as a region that benefited from the initial phase of implementation of REP. The target population comprised of households of electricity adopters and non-adopters and electricity utilities. Several sampling procedures were used in selection of the required respondents and locations. Multi stage random sampling procedure was employed in selection of divisions, and locations and sub- locations where households were to be interviewed. First stage sampling was used in the division of the Sub-County into three existing divisions. In second stage, simple random sampling was used to select an approximate target of half the number of locations in each division. In Chuka Division, three locations were selected

(Mugirirwa, Karingani and Mugwe locations), and in Magumoni Division three locations were also randomly selected which were; Thuita, Rubati and Kabuboni locations. Finally in Igamba Ng‘ombe Division two locations selected were Itugururu and Kamaindi. In this stage, a total of eight locations were selected from the divisions.

Similarly, two sub-locations were selected from each location bringing to a total of 16

sub-locations. Since the population in the study area varies the number of households interviewed was proportionally determined, relative to the number of households in each sub-location. This Information on the population was acquired from the District

Development Officer at Chuka, the district headquarters. Simple random sampling was used to obtain the households from each of the selected sub-locations. Every zone following a minor road or footpath leading to rural residences was considered to be a random way for selecting the sample. Therefore, every 10th household, to either the left or the right of a footpath or minor road, was chosen to be interviewed. A total of

150 households were selected from the district for the study.

The target population consisted a total of 33,293 households. The CRS (2009) software uses the following formula in the calculation of a sample size:

S= ------equation (1)

Where S=sample size, Z=Z value (1.96 for 95% Confidence level) p=percentage picking a choice, expressed as decimal (0.5) and c=confidence interval, expressed as decimal (0.08). This formula calculated the number of households for the study across the three divisions. A total of 150 households were selected. Transformer points in the selected sub-locations were recorded.

Figure 3.2: Sampling Strategy of Households in Meru-South Sub-County

MERU- SOUTH SUB-COUNTY (The Study Area)

Chuka Igamba Ng’ombe Magumoni

Mugwe Karingani Mariani Thuita Rubati Kabuboni Itugururu Kamaindi

H ) ) ) 7 1 ) 3 ) ( ) ) 1 ) 1 1 )

9 ( 9 ) 9 ( ) ) ) 1 ) 9 ( 6 ) (

u (

) ( ( 8 (

8 7 7

5 ( u i

i d a ( (

2 ( ( i

a d n 1 n n i

( i e w i a h r (

u i

n o u r w r

c b

k n a u i t i d i t m n o g e i u t r a a r a b a h a n g n m i g i i a t y h u h w n o m o g t e r m m k b g l t n n t ’ r a a u a u a a i i a i a a a a u u g u t g g K M R K K M K K R K K M I I I N K

Source: Author, 2014

3.7 Data Collection Tools

The questionnaire was the main instrument used to collect primary data from the households (Appendix I). It was divided into the following themes; Household demographic characteristics, electricity adoption status, electricity use patterns and benefits. The targeted respondents were the household head, and in their absence the spouse of the household head or a close and mature relative or next of kin was interviewed. The questionnaire combined open ended and closed ended questions.

The geographical coordinates of selected adopters and non-adopters households and transformer points, were recorded using hand held Garmin GPS sets for the purpose

of appropriate geo-referencing. This information was used to generate maps using

GIS technique that showed spatial distribution. Attribute information regarding the transformers in the study area were also gathered through consultations and discussions with electricity service providers and the questionnaire.

To complement the information gathered using the questionnaires, structured interviews were used to collect data from key informants from REA, KP and Local

Administration on electricity distribution and other developmental aspects within the sub-County (Appendix I and II). Secondary data was sought from annual reports of

Kenya Power, Ministry of Energy, and other relevant published reports.

3.8 Data Processing and Analysis

Data collected was first pre-processed where the questionnaires were examined and cleaned to ensure they were completed and consistently filled. Thereafter, the response questions were numerically coded and responses stored in a database template under assigned variable names for analysis using Statistical Package for

Social Sciences (SPSS) computer software version 19.

Data entered was subjected to Data View option to display in order to cross check the accuracy of data entered. This exercise led immediately to checking of errors and omissions, cleaning of errors and controlling data quality. This process was vital towards solid foundation for analytical processing of data both qualitatively and quantitatively. Thus, it is the bases for success in the final analysis. The qualitative data collected using the open ended questionnaire and interview guide, was taken in form of notes and used to compliment the quantitative data thematically.

3.8.1 Descriptive Statistics Analysis

Analysis of the socio-economic characteristics of the household heads, level of adoption and the socio-economic profiles of adopters and non-adopters was done using descriptive statistics such as frequencies, means, standard deviation, percentages. This analysis was performed using SPSS version 19 and Microsoft Excel was used to enable graphical representation. The study adopted Garson (2012) guideline that before starting with any advanced analysis, it is always good to start with some descriptive statistics and simple graphics.

3.8.2 Non-parametric Test

To examine the nature of the relationship between independent and dependent variables used in this study was conducted through a non-parametric tool, the chi- square. Chi-square test is a nonparametric statistical test to determine if the two or more classifications of the samples are independent or not (Zibran, 2012).

Thus the chi-square formula is:

k (Oi  Ei)2  2   ------Ei equation (2)

Where k = # of categories, Oi = observed number of cases in each category, Ei = expected number of cases in each category. In this study Chi square was was used to test whether the explanatory variables were related among the adopters and non- adopters. The results were then tested for significance at 0.05 (95 percent confidence level).

Independent sample t-test was used to assess whether or not the mean satisfaction scores of electrified and un-electrified public facilities are statistically different. This test provides mean difference, t-values, degrees of freedom and their significance.

According to Park, (2009) independent sample t-test is a statistical analysis tool used to compare mean scores on the continuous variable for two different independent groups. In this study t-test was used to detect difference in households‘ perception on quality service provision in electrified and non-electrified public facilities. Thus the t- test formula is:

1  2 t= , df = (n1-1) + (n2-1) ------equation (3)

Where is the difference in sample results and s is the difference in standard error (Park, 2009). The results were then tested for significance at 0.05 (95 percent confidence level).

In order to assess interaction effects between the independent variables and investigate the socio economic determinants of electricity adoption, independent variables were screened for multicollinearity using Colinearity Diagnostics function in SPSS (Leech, Barrett, & Morgan, 2005). Consequently, binary logistic regression analysis was performed to examine the influence of various household socio- economic characteristic on electricity adoption. This regression method was chosen because binary logistic regression is primarily used when the dependent variable is a dummy categorical variable (usually dichotomous) and has two outcomes such as 0 and 1. More so, logistic regression is often chosen when the predictor variables are a mix of continuous and categorical variables. Logistic regression makes no assumptions about the distributions of the predictor variables (Peng, 2002). In this

study, the dependent variable (adoption of electricity by households) is a dichotomous variable. Hence a value of 0 was assigned if the household was of a non- adopter and

1 if the household was an adopter, giving a regression of a non-linear form. This was done following a guide line provided by Hilbe (2009).

Thus, probability of adoption is explained as follows:

ln (p/1-p) =a + b1x1+b2x2+…+bnxn------equation (4)

The logit transformation of the probability of adoption is represented as follows:

Logit (p) = a + b1x1+b2x2+…+bnxn------equation (5)

Where p: is the probability of a case belonging to category 1, p/1-p: the odds of electricity adoption, a: constant, n: number of predictors and b1-bn represents regression coefficients.

The spatial distribution patterns of electricity accessibility points, adopters and non- adopters households mapping was done using GPS coordinates collected during fieldwork. The GPS coordinates were reorganized into a GIS compatible file and imported into ArCGIS 10.2 to generate maps showing location of adopters, non- adopters and electricity accessibility points in the study area. The summary of the various objectives, data types used, method adapted for the analysis with the corresponding statistical tools presented in (Table 3.1).

Table 3.1: Variables for Data Collection

Objectives Type of data collected Data collection Data analysis Dependent Independent variable Tools variable 1. Examine how Adoption Household characteristics; Questionnaire Descriptive household social- Age, Gender, marital statistics; economic status, education level, Percentages, Mean, characteristics occupation of the standard deviations, influence household head, sources Frequencies, cross electricity adoption of income, monthly tabulations. Chi- in the study area. income, and main type of square. Regression dwelling. (Binary logistic regression model), 2. Assess the socio- Electrification Electricity use Questionnaire Descriptive statistics; economic benefits benefits of various Percentages, mean, and challenges of activities standard deviations, electricity adoption frequencies. Chi- among households square. in the study area. 3. Evaluate the Electrification Quality of service Questionnaire Descriptive statistics; effect of rural delivery Percentages, mean, electrification on Interview guide standard deviations, development of frequencies and cross public facilities tabulation t-test 4. Examine the Adoption Accessibility GPS sets Using ArCGIS spatial distribution Questionnaire software version 10.2 of electricity layers created and adoption, non generated into maps. adoption and accessibility in the study area. Source: Author, 2013

3.9 Ethical Considerations

According to Hemmings (2006) it is important that a researcher embraces respect, confidentiality and sensitivities of their research participants and also the integrity of the institutions within which the research occurs and its research policy. The research proposal was approved by the research and Ethics Committee of Kenyatta University while the Ministry of Higher Education, Science and Technology gave consent to conduct the research. Permission was also obtained from Electricity sector utilities and local administration in the study area. Additionally, participant‗s right and privacy has been addressed as indicated in the preamble of the research questionnaire

and interview guide (Appendix I, II and II) in order to guarantee the participant‗s confidentiality and keep all responses anonymous throughout the study and thereafter.

3.10 Interpretation and Presentation of Results

Subsequently after the data was analyzed through cross tabulation, frequencies, percentages, chi-square, t-test and binary logistic regression model, the results were explained in form of frequencies, percentages and displayed in tables, pie charts, graphs photographs and maps and text form for easy understanding. Descriptive results were captured into MS Excel to enable their graphical representation using cross-tabulations and charts. The findings were triangulated with work done by scholars in the area of electricity adoption and access in rural areas in general to assess the relevance of the findings made. This procedure aided in drawing meaningful and comprehensive conclusions from the results obtained. The results and discussion, summary of the findings, conclusions and recommendations are presented in the subsequent chapters of the study. The Statistical Package for Social Sciences was used to analyse the data.

CHAPTER FOUR

RESULTS AND DISCUSSISON

4.0 Introduction

This chapter presents the results and discussions in tandem with the study objectives.

The results presented (in tabulation, graphics or in figures) are described followed by a discussion with their analyses and literature re-evaluation which subsequently forms the basis of the conclusions drawn. Results on determinants of electricity adoption are discussed first followed by effects of electricity on households and public facilities, and finally spatial distribution of transformers, adopters and non-adopters.

4.1 Socio-economic Characteristics of the Respondents

This study was carried out in Meru-South Sub County in Tharaka Nithi County, where a total of 150 households were interviewed. The respondents interviewed consisted of 97% and 3%, of household heads and spouses respectively, primarily indicating the reliability of the responses as it were assumed that these members have accurate facts about the status of the household activities. Marital status of household head indicated that 91.3% were married, 4.7% were widowed and 4% were single parents. For the entire sampled households 94% and 6% constituted male and female headed households respectively (Table 4.1). In levels of education attainment of the household heads, 5.3%, 40%, 41.3%, and 13.3% for non-formal education, primary level, secondary level and tertiary level respectively. As per the household‘s heads occupation, 18.7%, 73.3%, 2.7%, and 5.3% were employed, self-employed, daily labourer and unemployed, respectively.

Table 4.1: Summary of Socio-economic Characteristics of the Household Heads

N=150 Variables Percentage (%) Gender Male 94 Female 6 Marital status Married 91.3 widowed 4.7 Single parents 2.0 Single 2.0 Education No formal 5.3 education Primary 40.0 Secondary 41.3 Tertiary 13.3 Occupation Employed 18.7 Self employed 73.3 Daily laborer 2.7 Unemployed 5.3 Main Source of income Salary 19.3 Farming 68.0 Wages 4.0 Business 8.7 Monthly Income (Kshs.) 5000 and 30.7 below 5001-10000 39.3 10001-20000 17.3 20001-30000 6.0 Above 30001 6.7 House Wall material Wood 52.0 Stone 24.0 Mud 12.7 Brick 10.0 Iron sheet 7.0 Age (years) 25-35 12.7 36-45 28.0 46-55 33.3 56-65 19.3 Above 65 6.7 ≤5000 30.7 Monthly income 5,001-10,000 39.3 10,001-20,000 17.3 20001-30,000 6.0 >30,000 6.7 Distance of the household ≤ 600 metres 40 from the nearest transformer > 600 metres 60 Source: Field data, 2013

The study results indicate a distribution of ages of the heads of households interviewed ranging between 25 to 83 years, and the average age of the household head being 49 years. The study further revealed that 36.7% were aged between 25 and

45 years, 56.7% between 46 and 65, giving a cumulative figure of about 93.3% of household heads interviewed aged between 25 and 65 years. The average age of household heads was 48.74, median 48.50 and the mode 48 years.

4.2 Household Socio-economic Characteristics Influencing Electricity

Adoption in Meru-South Sub-County

The first objective of the study was to find out which household socio-economic factors influenced electricity adoption as the characteristic may influence decision making in adoption of electricity. The study laid emphasis on level of education, income of the household, employment status, family size, and age of the respondent, distance from the transformer, house type as well as gender of the respondent. This section provides descriptive statistics of socio-economic characteristics of adopters and non-adopters households and logistic regression analysis was used to predict the determinants of electricity adoption among households.

4.2.1 Status of Electricity Adoption

Among the respondent households, 36% were adopters, while 64% were non- adopters. According to the results, electricity adoption among households was low in

Meru-South Sub-County.

4.2.2 Gender Characteristics of Adopter and Non-Adopter Household Heads

Results revealed that 92.7% of non-adopters were male headed households and 7.3% were female headed households. Additionally, 96.3% and 3.7% of males and females households were respectively adopters (Table 4.2).

Table 4.2: Gender Characteristics of Electricity Adopters and Non-Adopters

Gender Non- adopters Adopters Total No % No % No % Male 89 92.7 52 96.3 141 94.5 Female 7 7.3 2 3.7 9 5.5 Total 96 100 54 100 150 100 Source: Field data, 2013

Results showed that males were majority adopters. Male-headed household are more likely to get connected to electricity as compared to the female headed households

(Nishimwe et al. 2014). According to Clancy, (2003) males initiate and manage most household activities, among them energy. According to Ambunda et al. (2008) the traditional patriarchal system confers most rural men as household decision-makers.

4.2.3 Age Characteristics of Adopters and Non-Adopters Household Heads

From the results, majority adopters (37%) and non-adopters (31.3%) were aged between 46 and 55 years old (Figure 4.1). The similarity in modal class of adopters and non-adopters household heads may be due to the reason that overall, the majority of the household heads are under the same age bracket (46-55). In addition, the results showed that 33.3% of adopters were in the age bracket of 36 to 55 and a few in the older and the younger ages, while non- adopters took a percentage lead in the youngest and the oldest age brackets (25-35 and above 65).

Figure 4.1 Age Distribution of Adopters and Non-Adopters

40.00%

35.00% 30.00% 25.00% Non adopters 20.00% 15.00% Adopters 10.00%

5.00% Percentage respondents of Percentage 0.00% 25-35 36-45 46-55 56-65 above 65 Age in years

Source: Field data, 2013

The chi square test results showed no significant difference (p=0.725 at<0.05 probability) among adopters and non-adopters across the different age categories.

However, it was evident that adoption was high among the middle age brackets (36-

45 and 46-55) compared to their counterpart in the younger (25-35) and the older

(above 65) categories. This pattern is due to the fact that household heads in the younger age bracket have smaller families and less diverse uses of electricity, while very older people might have no interest in adopting modern sources of energy connection as they have lived without it over a long life span (Jan, 2012).

4.2.4 Educational Characteristics of Adopters and Non- Adopters

Educational attainment is one of the key factors in terms of decision on whether to connect to grid electricity especially, in the rural areas. Study findings illustrate a distinction among the two categories. The percentage of adopters increased among

households with higher education and decreased in among households with no formal education and lower levels of education (Figure 4.2).

Figure 4.2: Educational Levels among Electricity Adopters and Non- Adopters

60.00% 50.00% 40.00% 30.00% Non adopters 20.00%

10.00% Adopters Percentage respondents of Percentage 0.00% none primary secondary tertiary

Education status

Source: Field data, 2013

This findings contrast with that of non-adopters whose percentage decreases in higher education and increases with no formal education and lower levels of education.

Conclusively, the study established that majority of household heads in higher levels of education are adopters whereas non-adopters are mainly in no formal education and primary education level.

A Chi-square test to compare the educational characteristics of adopters and non- adopters indicated that there was a significant (p=0.000 at probability<0.05) difference between education levels of adopters and non-adopters. This suggests that the decision on whether to adopt or not adopt electricity may be influenced by education level of household head. 37

A cross tabulation of age with level of education of adopter and non-adopter household heads indicated that majority (90.0%) of respondents in lower education levels were in age bracket (>65), where none (0.0%) had attained secondary and tertiary education (Table 4.3).

Table 4.3: Link between Education and Age Status of Adopters and Non-

Adopters

age class (years) non- adopters age class (years) adopters Percent % 25-45 46-65 >65 25-45 46-65 >65 None or primary 82.6 78.3 90 17.4 21.7 10 Secondary 51.7 54.5 0 48.3 45.5 0 Tertiary 41.7 1.1 0 58.3 88.8 0 Total 40.6 50 9.4 40.7 57.4 1.9 Source: Field data, 2013

This low level of education among the older household heads could be due to lack of availability and zeal for schooling in the past. Parents might have considered that sending children to school was like reduction of labour in the homestead (Horgan,

2007).

4.2.5 Occupational Characteristics of the Household Heads

The results showed differences in occupational characteristics of the households. With respect to household heads who had adopted electricity, findings showed that 37.0% were employed (teachers, nurses, and clerks); 51.9 % were self-employed (owned small businesses and farmers); 9.3 % were unemployed and 1.9 % were daily laborers

(casual laborers in farms who earned an average of Kshs.200 per day) (Figure 4.3).

Figure 4.3: Occupational Characteristics of Household Heads

90.00% 80.00% 70.00% 60.00% 50.00% Non adopters 40.00% 30.00% Adopters Percentage respondents of Percentage 20.00% 10.00% 0.00% employed self daily laborer unemployed employed Occupation

Source: Field data, 2013

Employment characteristics of the 96 non-adopters revealed that 86.5% (83) were self-employed, 8.3% (8) were employed, 3.1 % were daily laborers (3) and 2.1% (2) were unemployed. Findings showed that those employed formed majority category in the adopters while non-adopters were the minority. This being typically a rural area, majority of the employed household heads either work in the available government institutions especially the schools, hospitals and government offices, hence more aware of the benefits of electricity. Among other occupations self-employed category had the majority of adopters (86.5%) and non-adopters (59.1%) who are mainly involved in agri-businesses and other forms of micro-enterprises.

4.2.6 Household Size of Adopters and Non-Adopters

Household size is a key factor which underlies adoption patterns of electricity in the rural areas. Evidence from the survey results shows a discrete difference with respect to adopters and non- adopters (Table 4.4). The household family size ranged from 2 to

12 members. Family size among the adopters and non-adopters was quite distinct. The survey revealed similar modal class of 5 members in both categories but with response rate of 31.6% and 5.8% among adopters and non-adopters respectively.

Table 4.4: Household Size Characteristics of Adopters and Non-Adopters

Non- Adopters Adopters Total Household size n Percentage (%) n Percentage (%) n Percentage (%) 1 0 0 0 0 0 0 2 2 25.0 6 75.0 8 100.0 3 13 48.1 14 51.9 27 100.0 4 23 71.9 9 28.1 32 100.0 5 31 67.4 15 32.6 46 100.0 6 > 27 73.0 10 27.0 37 100.0 Source: Field data, 2013

According to the results, majority of adopters (75%) had lesser household size as compared to the non-adopters (73%) who had a higher household size. In relation to household size, x2 test showed that there was a significant (p= 0.031 at <0.05) difference between the adopters and non-adopters based on the household size.

4.2.7 Marital Status of Adopters and Non-Adopters’ Household Heads

For adopters the percentage of married, widowed, single, and single parent stood at

88.9 %, 5.6 %, 1.9 % and 3.7 %, respectively. Non-adopters‘ category had almost similar percentages at 92.7%, 4.2%, 2.1 % and 1.% for married widowed, single and

single parent, respectively. Evidently, majority of the household heads in both groups were married and lived with their spouse (Figure 4.4).

Figure 4.4: Marital Status of Adopters and Non-Adopters

100.00%

90.00% 80.00% 70.00% Non adopters 60.00% 50.00% 40.00% Adopters 30.00% Percentage respondents of Percentage 20.00% 10.00% 0.00% single married single widowed parent Marital status

Source: Field data, 2013

4.2.8 Characteristics of Household’s Dwelling House by Type of Wall Material

According to the study results, all households surveyed owned their housing units.

This is a common occurrence in rural areas as majority of the population own land hence have put up houses. Essentially, house walling material, is an important factor in considering whether to adopt electricity or not among households. Some housing materials such as cartons, plastic bags and mud are not convenient especially when it comes to wiring and installation of bulbs. Apparently, the types of building materials differ among the respondents. Among the adopters a majority (48.1%) had stone walled houses, followed by timber walled houses (42.2%), while 3.7% had mud

walled and iron sheet walled houses and the rest (1.9%) lived in brick walled houses.

Among the non-adopters majority (57.30%) had timber walled houses, 17.7% had mud walled houses, followed by 14.6% who had brick houses and finally a 10.4% who had stone walled houses (Figure 4.5).

Figure 4.5: Household Distribution by Wall Type

60.00%

50.00%

40.00%

30.00% Non adopters

20.00%

Adopters Percentage respondents of Percentage 10.00%

0.00% brick stone mud wood iron sheet House type

Source: Field data, 2013

According to the survey results, house wall type differed among the adopter and non- adopter households. The most dominant wall material amongst the adopters and non- adopters was stone and wood at 47.4% and 58.1 %, respectively.

4.2.9 Main Sources of Income of the Household Heads

Main sources of income differed among the adopters and non-adopters. Among the adopters their main sources of income was at 53.7%, 29.6%, 16.7% and 0.0 % from

farming, salary, business and wages respectively. More so, the pattern of income sources among the non-adopters pattern was as follows; 80.2 % was from farming, 5.2

% from salary, 5.2 % from wages and 9.4 % from business (Figure 4.6).

Figure 4.6: Sources of Income among Adopters and Non-Adopters

90.00% 80.00% 70.00% 60.00% 50.00% Non adopters 40.00% Adopters 30.00%

Percentage respondents of Percentage 20.00% 10.00% 0.00% salary farming wages bussiness Sources of income

Source: Field data, 2013

Household income among the electricity adopters is mainly derived from salary followed by small scale farming activities. Consequently non-adopters had the highest proportion of income from small scale farming activities and minority (5.2%) earning their income from wages. The results further revealed that despite majority of adopter household heads being employed about 7.4% had farming as their main source. This implies that farming may be a better source of income even for persons who are employed.

4.2.10 Monthly Total Income of Adopters and Non-Adopters

According to the results (Table 4.5) there was disparity between the adopters and non- adopters of electricity based on monthly total income. The majority at 44.4% of the adopter household fell in the income bracket of Kshs. 5,000 to 10,000 closely followed by 27.8% in the income bracket of 10,001 to 20,000. Additionally, majority

(45.8%) of non-adopters were in the least income group of less than 5,000 as compared to adopters household head who dominated in the highest income bracket of 30,001 and above (Table 4.5).

Table 4.5: Monthly Total Income of Adopters and Non- Adopters

Non-adopters Adopters Average monthly income Frequency (n) Percentage (%) Frequency (n) Percentage (%) Less than 5000 44 45.8 2 3.7 5,000-10,000 36 37.5 24 44.4 10,001-20,000 10 10.4 15 27.8 20,001-30,000 5 5.2 4 7.4 More than 30,001 1 1 9 16.7 Total 96 100 54 100 Source: Field data, 2013

According to the results, 45.8% and 3.7% of adopters and non-adopters respectively earned less than Kshs. 5,000. The second income category of Kshs. 5,001-10,000 comprised 37.5% adopters and 44.4% of non-adopters. Income group of Kshs.

10,001-20,000 comprised 10.4% and 27.8% of adopters and non-adopters respectively. Income group Kshs. 20,001-30,000 comprised 5.2% of adopters and

7.4% of non-adopters. The highest income group comprised 1.0% of non-adopters and

16.7% of adopters. These findings revealed that the majority adopters are in higher income groups of whereas the non-adopters are in the lowest income category.

4.3 Extent to which Households’ Socio-Economic Characteristics Influence

Electricity Adoption

The first research hypothesis of this study focused on the direction and magnitude of relationships between each independent (socio-economic characteristics) and a binary dependent (electricity adoption and non-adoption) variable. Before the logistic regression was conducted, the independent variables were screened for multicollinearity.

Table 4.6: Test for Multicollinearity

Colinearity Statistics Model variables Tolerance VIF* Distance from the transformer .884 1.131 Gender .824 1.213 Age .847 1.180 Marital status .840 1.190 Education attainment .558 1.791 Occupation .708 1.412 Family size .893 1.119 House type .928 1.078 Main source of income .871 1.148 Monthly income .555 1.800 *Variation Inflation Factor

Source: Field data, 2013

Hair et al., (2010) suggest Variation Inflation Factor value larger than 1.0 or tolerance value smaller than 0.10 indicates the problem of multicollinearity. It is observed that the VIF factors for all the nine variables were within 2.0 (Table 4.6). Therefore, all the nine variables were included in the logistic regression analysis. The binary logistic regression function in SPSS was used to conduct the simultaneous logistic regression analysis and results are presented in (Table 4.7). The model was significant (χ2 =

118.558, df= 20, p < 0.05) and correctly predicted 88.7% of both adopters and non- adopters.

In this study, distance from the transformer, education level, gender, household size, and monthly income, were found to be significant (p=<0.05), and were positive predictors that influenced electricity adoption, hence leading to rejecting the null hypothesis that there is no significant relationship between households‘ socio- economic characteristics and electricity adoption in Meru-South Sub-County. Other four variables that were found not to be significant include; occupation, source of income, house type, marital status, and age.

Table 4.7: Logistic Regression Analysis of Explanatory Variables in Electricity

Adoption

Variables B S.E. Wald Sig Exp(B) Gender -2.774 1.386 4.004 0.045* 0.062 Age 0.036 0.254 0.020 0.887 1.037 Marital status 1.086 0.575 3.561 0.059 2.963 Education level 1.001 0.429 5.439 0.020* 2.720 Occupation 0.612 0.499 1.506 0.220 1.844 Household size -0.589 0.226 6.789 0.009* 0.555 House type 0.240 0.240 0.999 0.318 1.271 Main source of income 0.293 0.309 0.898 0.343 0.746 Household income 0.824 0.324 6.458 0.011* 2.281 Distance from the transformer 3.167 0.569 31.013 0.000* 0.042 N=150, *Significant at 5% probability level B, Parameter estimate; SE, Standard error. -2log likelihood is 77.37; Chi square statistic is 118.558*; Overall correct prediction is 88.7%. Source: Field data, 2013

4.3.1 Distance from the Transformer

Distance of the household form the transformer was statistically significant p=0.000 at <0.005. A positive sign of beta (β=+3.167) indicates that adoption of electricity is negatively influenced by the distance of the household to the transformer. The results imply that the longer distance of the household from the transformer the less likely that the households will adopt electricity. Therefore households within 0.6 kilometres from the transformer are more likely to connect electricity to their households in comparison to those households not within 0.6 kilometre radius, keeping other variables in the model constant. According to Mapako et al. (2007) households near grid electrified growth points usually benefit from the proximity of the grid hence get easily connected. This finding is consistent with that of Kembo (2013) who noted that

93.0% of the electricity connections were within a radius of two kilometers from the main electricity grid as well as the tarmac roads, hence rural communities and households with easy access to tarmac road and closer to the grid were connected to the grid. Evidently, electricity accessibility is an indicator of electricity adoption patterns. In spite of the power distribution model, proximity from power supply is a vital aspect influencing electricity adoption by households in rural areas. In essence, an econometric analyses conducted in India demonstrated that the location of a village from the grid is a key determinant of electricity connection by households therefore, villages closer to the grid are easily connected for technical and financial reasons compared to areas far from the grid (Andreas 2006). According to Chakrabati and

Chakrabati (2002) the cost of electricity grid supply via grid extension increases considerably as the distance from the grid to the village increases. Findings on this issue can have practical implications on the future planning hence; this hypothesis is worth further exploration. More so, Andreas (2006) indicated that increase in

neighborhood electrification by 1.0% increases the probability of household access by

1.0%.

4.3.2 Education attainment

In previous studies, the relation between level of education and electricity adoption and non-adoption has seldom been tested. It is uncertain whether electricity adoption is associated with school attendance. According to the results, education level has a significant (p=0.020) and positive (β=1.001) influence on electricity adoption. This implies that households with household heads with higher levels of education are likely to be electricity adopters. These results points out that the higher the level of education the higher the likelihood of electricity adoption. With formal education households have a better understanding of the importance of electricity in their households. This study concurs with findings by Olufemi et al. (2012) who stated that electricity adoption and use are positively associated with higher level of education.

Therefore, household heads who are highly educated especially in rural areas have better preferences especially when it comes to choosing clean energy sources such as electricity because they are conversant with the importance of its use. This finding also concurs with that of Andreas (2006) that the probability of electricity connection increases considerably as the education levels of household head rise. According to a study by Hing et al. (2012) the educational levels of respondents that are connected to grid electricity mainly have acquired secondary and tertiary education. Khandker et al. (2009) further stated that education opens up opportunities for wage employment and other economic activities outside farming, therefore providing extra income which is used in acquiring modern energy sources such as electricity among other necessities.

4.3.3 Income

Household income was found to be a significant at (p=0.011 at <0.05) in electricity adoption. A positive sign of beta (β=0.824) indicates that electricity adoption is positively influenced by the household monthly income. This could be because higher income earners have finances to facilitate electricity connection to their houses. This research finding suggests that increase in household income increases the likelihood of adopting electricity. The results agree with those of Heltberg (2003) that low income households are less likely to adopt electricity due to the high initial cost of connecting to the grid, which includes the infrastructure, cost (obtaining connection to the grid) as well as the cost of obtaining appliances to utilize with the service. Results are in consensus with findings by Barnes (2007) that initial connection charges still remains a challenge when it comes to electricity adoption among low income earners households. According to Mills and Schleich (2010) richer households are less likely to face income or credit constraints for investments in modern energy sources such as electricity.

4.3.4 Household Size

The study found out that household size was a significant (p=0.009 at <0.05) predictor. Further, the predicted coefficient (β=-0.589) of household size was negative hence having a negative relationship. This implies that households with fewer members were more likely to be electricity adopters compared to households with more members. According to Pachauri (2007) household size is a determinant factor in electricity adoption among households. As noted by Mekonnen and Köhlin (2009) households‘ characteristics such as family size is significant in determining whether household adopts electricity.

4.3.5 Gender

Gender was statistically significant p=0.045 with a predicted coefficients of -2.774 indicating a negative relationship. This suggests that implied that male headed households are likely to be electricity adopters than are their female counterparts. This result is consistent with a study by Dreze and Srinivasan (1997), which found out that more male-headed household had electricity connected to their houses compared to the female headed houses. A study carried out in South Africa found out that female- headed households were consistently less likely to be connected to the main source of electricity than those headed by males (StatsSA, 2010). According to Nishimwe et al.

(2014) female-headed households appear to be less likely to secure electricity for their homes compared to those headed by males. A study by Dungumaro (2008) revealed that households headed by females are relatively underprivileged in terms of assets and income or are significantly over-represented among the poor.

From the binary logistic regression analysis, it is evident that distances from the transformer, education level, gender, household size, income were main factors influencing electricity adoption in the area. Effort to increase adoption should first address these issues in order to increase adoption among households as they are focal points for rural development.

4.4 Socio-economic Benefits and Challenge, of Rural Electrification

Adoption among Households

This section assesses the social and economic beneficial aspects of electricity adoption and households‘ perceptions on assorted adoption challenges. When accessing socio-economic benefits of electricity it is important to first understand the period that households have been connected to the grid electricity as it provides background information for various interventions geared towards socio-economic development. According to the findings, households had been connected to grid electricity for periods ranging from 1 to 10 years. The average was 3.7 years, and the median was 4 years. Evidently, years prior rural electrification inception (2009) in

Meru-South have presented low connection level with recent years having a higher number of households getting connected to the electricity (Table 4.8)

Table 4.8: Number of Years Households Connected to Grid Electricity

Years Frequency Percentage (%) Cumulative Percent (%) 1 8 14.8 14.8 2 10 18.5 33.3 3 7 13 46.3 4 18 33.3 79.6 5 4 7.4 87 6 2 3.7 90.7 7 1 1.9 92.6 8 1 1.9 94.5 10 3 5.6 100 Total 54 54 100 100 Source: Field data, 2013

As the results indicate, a cumulative of 87% of households had connected between

2009 and 2013. The results suggest that due to accelerated accessibility to electricity in the last five years (Figure 4.7) where rural electrification saw unprecedented

increase in electrification of public facilities and increased connection among households. Prior to 2009 when the rural electrification programme had been newly initiated, only a few households (9.3%) had access to the grid hence fewer existing connections among households.

Figure 4.7: Households Connected to Electricity in Year 2009-2012

35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00%

Percentage householdsof Percentage 0.00% 2009 2010 2011 2012 2013 Years

Source: Field data, 2013

Yet again, this finding agrees with a study by IEG, (2008) which indicate that when electricity has been provided to a community, majority of households that connect do so in the first three years that the grid is available. The decline in the year 2011 was mainly associated with inadequate funding for the exercise. This drift was well elaborated by the Kenya Power officials who explained the financial difficulties experienced through this duration. It was until mid 2011 when funds were availed and the process recommenced and led to further distribution of electricity and connection in the study area.

4.4.1 Uses of Electricity among Households

Study findings revealed ways in which electricity was utilized in the household, based on the socio-economic productive uses of electricity. Apparently, there was diverse application or uses of electricity among the households which were identified from the survey findings. First, the study sought to understand appliance ownership among the adopters because as cited by Thom (2000), majority of households do acquire electric equipment/appliance after electricity is available in their households.

Lighting devices were the dominant in all households that had been connected households at 100%, and consequently households used electricity for space illumination as the prime use of electricity (Table 4.9). The second most owned electric appliances were radios at 96.3% and television sets at 94.4%. Other electric appliances included electric iron and refrigerator, whose ownership was at 27.8%, and

9.3%, respectively among sampled adopter households.

Table 4.9: Electric Appliance Ownership among Electricity Adopters

Appliances Frequency Percent of Cases SRadio 52 96.3 oTelevision set 51 94.4 uRefrigerator 5 9.3 r Electric iron 15 27.8 c Electric heater 6 11.1 SMobile phone 53 98.1 oComputer 2 3.7 uLighting devices 54 100.0 r Chaff cutter 15 27.8 c Total 253 468.6% e Source: Field data, 2013

The results concurs with a study by Wamukonya et al. (2001) who noted that refrigerator is a luxury item owned by a few especially in rural areas and its ownership which is heavily dependent on households‘ income.

Mobile phones were also owned by majority of adopter households at 98.1% indicating ease of charging the mobile phones. Electric appliances used in increasing productivity in agriculture were not prevalent among the households where only

27.8% owned chaff cutters used in cutting fodder. Majority adopters of the households did not have equipment necessary for agricultural productivity, a paradox for an area that is a highly productive agricultural region. This finding differs with the scenario in India where agricultural productive areas had been supplied with appropriate equipment such as irrigation pumps in order to enhance agricultural productivity (Badiani, 2011). Electric heater ownership among the households stood at 11.1% and computer ownership at 3.7% as many of the households could not afford to buy or did not use minigargets in their homes.

Electricity is hardly in demand for itself, but for the outputs derived from the use of various electric appliances. Once electricity has been generated and distributed, the final output is appliance driven. Respondents who owned various electric appliances were asked to state various electricity driven output from the appliances as represented in (Table 4.10).

Table 4.10: Electricity Benefits from Appliances in the Households

Appliances Use Frequency Percent % Radio1 Access to information 48 88.9 Radio2 Entertainment 4 7.4 Television1 Entertainment 44 81.5 Television2 Access to information 43 79.6 Preservation of food and Refrigerator beverages 5 9.3 Electric iron Convenience 15 27.8 Mobile phone benefit Improved communication 53 98.1 Computer Access to information 2 3.7 Lighting device 1 Improved lighting of spaces 54 100 Lighting device 2 Improved security 15 27.8 Chaff cutter Increased productivity 15 27.8 Source: Field data, 2013 1first priority use 2second priority use

The findings showed that about 100% of all connected households owned lighting devices which included fluorescent tubes, incandescent light bulbs and energy saving compact fluorescent light bulbs which were used in lighting spaces. Lighting was the first priority for being connected to the grid electricity.

This finding concur with that of Chaurey et al. (2004) whose study found out that the initial use of electricity in rural areas is household lighting because electric light is much brighter than that provided by kerosene lamps and the price per unit of light can be hundreds of times cheaper. Apart from indoor illumination about 27.8% of the households used electricity for security lighting at various points in the homestead especially at the main entrance (gate) to the residence.

About 96.3% of the adopters owned radios and the dominant adopters at 88.9% used them mainly in accessing information. This can be explained by reasons such as radios are cheaper to purchase, the availability of local stations that broadcast in vernacular languages; availability of local news and educational programmes especially for farmers. Moreover 7.4% mainly used radios for entertainment purposes, listening to entertainment programs such as music jams storytelling and audio dramas.

Mobile phones were owned by 98.1% of adopters who reported that with electricity connection it became more convenient for them to use the device. This improved communication with family members in distant places, money transfers, social media and internet. Availability of power for charging the cell phones had made it very convenient to households. Television sets were also a common electric appliance among households whose major output was for entertainment as reported by 81.5% of

households and access to information at 79.6%. This finding is almost similar to a study by Barnes et al. (2005) that indicated the second most common use of electricity is television and on average, close to half of all electrified homes in rural

India have a television and use it for entertainment. It was noteworthy that prices of television sets have fallen thus making them available to many households.

Several other electrical appliances were owned by a smaller proportion of households, in particular, the refrigerator, that households used by only 9.3% in preservation of food and beverages. Residents with refrigerators reported that they did not have to worry about wasting perishable food though at times long periods of unplanned power outages resulted in large quantities of spoiled food. Computers were used by only

3.7% of the households who mainly used it to access information. Chaff cutter agricultural equipment was owned by 27.8% of adopters and was reported to having increased agricultural productivity among the users. Chaff cutters were used by farmers who kept livestock and majority who practiced zero grazing.

Electric iron use was reported to be at 27.8% of the adopters and was used as a convenient device for ironing clothes although, ironing clothes was not regarded as a priority and the majority (53.3%) did not use it frequently as shown in (Table 4.11).

Table 1.11: Frequency Use of Electric Appliances

Appliance Frequency used Frequency Percentage Electric iron Daily 2 13.3 Weekly 5 33.3 Monthly 8 53.3 Water heater Daily 2 13.3 Weekly 1 16.7 Monthly 3 50 Source: Field data, 2013

Primarily, electrical appliances are used throughout the day to make their daily tasks and chores easier and improve quality of life. However, this was not the case in the study area as respondents with ironing box did not regard ironing clothes as priority as the majority did not use it frequently.

The electric heater, another electrical device only categorized as a cooking device was used by only 11.1% of adopters, and then infrequently used. This finding concurs with a study carried out by Bhattacharyya (2006) in India that stated that even with grid electricity, it is rare for electricity to be used for cooking other than heating small quantities of water for bathing or making beverages. However the findings of this study differ with that of Thom, (2000) in South Africa that stated that a significant percentage of rural households used electricity for cooking, although only among higher income households. Various reasons accounted for the low use in cooking including the cost of consumption as households are aware of the rapidly spinning wheel of the electricity meter if a heating ring is turned on. A few respondents in this study claimed they preferred the taste of food cooked over wood or charcoal, while believing that cooking with electricity is said to be dangerous, especially with a poor wiring. Hence, there is some resistance to using electricity for cooking, partly economic such as high cost of cooking using electricity and partly social Overcoming that resistance requires consumer education on proper use and handling of electricity to avoid misconceptions on electricity and subsidize electricity cost.

In this study, majority of households with electricity did not own equipment for increasing agricultural productivity apart from chaff cutters. This concurs with

Schelling, (2007) that most households especially those practicing small scale

agricultural do not acquire electricity for their farming. However, it is noted that electricity provides power for farm machinery, such as water pumps, threshers, grinders, and dryers. Farmers‘ acquisition of electric farming equipment would result in the modernization of agricultural production (Kopp & Mas, 2010). The cost and availability of electric appliances, has often been a prohibiting factor in the take up of electricity. Appliance costs should be subsidized and be locally available to increase the demand for electricity takes up and use among households.

4.4.2 Electricity Benefits from Home Business

The study sought to establish the benefits that households connected to electricity gained from the use of electricity with a focus on home business opportunities and general appreciation of the quality of life. According to the findings, only 18.0% (27) of the respondents run a business from the home. Out of this small number 77.8%

(21) were adopters and 22.2% (6) were non- adopters. Findings revealed that 61.1% of households did not carry out any business requiring electricity, whereas 14.8% had at least one business activity at home. Amongst those that declared having a business activity, 38.1% confirmed only one small business activity; 42.9% reported having two business activities and 19% reported having three businesses. This concurs with finding by Maleko (2005) that with availability of electricity there is diversification of business activities within the same household. Multiple studies have shown that microenterprise development is stimulated by electrification, even though other elements (such as availability of microfinance and organized local markets) are necessary to ensure that the RE has the desired impact (Kooijman et al., 2010; Bose et al., 2013; Maleko, 2005).

Household members were involved in several conventional small businesses that used electricity which included mobile phone charging at 44.74% as the most prevalent

(Figure 4.8).

Figure 4.8: Distribution of Small Business Activities in Households

10.53%

10.53% 44.74% Charging phone General shop Hair dressing salon Barber shop

34.21%

Source: Field data, 2013

Mobile phone charging was especially for those households that had not been connected with electricity and household members charged Kshs. 20. Other home businesses done were hair dressing and barber shops at 10.53% each, and general shop business at 34.21% selling electronics, food and hardware.

Home-based businesses provided income for the households. The households were asked to state their average income on a monthly basis from the small businesses carried out. Households with salon business reported an average monthly total income of Kshs. 6,000 as indicated in (Figure 4.9).

Figure 4.9: Average Income Generated from the Small Business

16,000 14,000 12,000 Mean 10,000 monthly 8,000 income 6,000 (Ksh)

percentage respondent percentage 4,000 2,000 salon hair cut charging general Shop phone Business Source: Field data, 2013

The barber businesses reported a mean monthly income of Kshs. 5,250, mobile phone charging with a mean monthly income of Kshs. 8,500, while general shop business reported a monthly income of Kshs. 14,769, hence having the highest mean monthly income among all small businesses conducted by households.

Utilization of income from the small business activities was revealed through various expenditure lines. Based on responses as presented in (Figure 4.10), three precedence expenditure items were noted; paying school fees for the children was the primary expenditure item at 86%, an indication that children‘s education is an important aspect of these households.

Figure 4.10: Uses of Income Generated from Small Business

90% 80% 70% 60% 50% 40% 30%

Percentage respondents of Percentage 20% 10% 0% Paying school fees Paying bills Domestic use Expenditure line items

Source: Field data, 2013

Paying electricity bills was reported by 19%. As people who depended wholly on electricity services, paying electricity bills meant ensuring the sustainability of the small business and consequently the continuous improvement of the welfare of the family. Domestic use of electricity was at 38% which included, purchasing electrical appliances, transport costs, household furniture, health matters and leisure for the family members.

4.4.3 Ranking of the welfare benefits of electricity

The study‘s results show a positive attitude on electricity among households towards children‘s education. Respondents believed that electricity had positive effects on their children‘s study time and consequently, with good implications for their education. They were asked to agree or disagree with the statement ―Having electricity is important for children‘s education‖. Results showed that 97.9% of all

non-adopters households strongly agreed with the statement and 100% of adopters strongly agreed (Table 4.12).

Table 4.12: Importance of Electricity to Children’s Education

Non-adopters Adopters

Responses % Responses % Rank Strongly agree 94 97.9 54 100.0 electricity is Agree 2 2.1 0 0.0 important for Don‘t know 0 0.0 0 0.0 children Disagree 0 0.0 0 0.0 education. Strongly disagree 0 0.0 0 0.0

Source: Field data, 2013

These results suggest that education among children is important and access to electricity improves the learning conditions hence, better education outcomes. More so, it is a significant factor in the development not only of children but also, communities as it helps in breaking chains of poverty. This is in agreement with IEA

(2008) who noted that rural electrification may affect education not only by improving the quality of schools resulting from their use of electricity-dependent equipment but also by increasing time allocation for studying at home.

4.4.4 Challenges in Electricity Adoption in Meru-South Sub-County

In order to determine the extent to which various aspects constrained electricity adoption, both adopters and non-adopters were asked to rate their perceptions on how selected factors challenged electricity connection to their households. This was on a

5-point Likert scale: +5 (strongly agree) to +1 (strongly disagree).

The study revealed that 72.0% of non-adopter households strongly agreed with the statement that the distance from the transformer was a challenge when it comes to electricity connection from the grid to their households. Another 18.51% of adopters

agreed that distance from the transformer was a challenge to grid connection (Table

4.13).

Table 4.13: Household’s Perceptions on Challenges of Electricity Adoption

Challenges Ranking Adopters Non-adopters Distance from the Strongly Agree 0.0% 72.0% transformer is a challenge Agree 18.5% 0.0% in connecting electricity to Strongly disagree 81.49% 27.8% my household Initial cost of electricity Strongly agree 16.7% 55.56% connection fee is a Agree 0% 17.845 challenge to electricity Disagree 83.3% 22.2% adoption Internal wiring cost Strongly agree 3.7% 5.2% challenges electricity Agree 0.0% 1.0% adoption in my household. Disagree 85.2% 92.7% Strongly disagree 11.1% 1.0% Information on electricity Strongly agree 13.0% 24.0% services is a challenge to Agree Disagree 7.4% 16.7% electricity adoption Strongly disagree 70.4% 55.2% 9.3% 4.2% Paying of electricity bills Disagree 88.5% 87.0% is challenge in electricity Strongly disagree 11.5% 13.0% adoption Source: Field data, 2013

Additionally 81.49% of adopters and 27.8% of non-adopters strongly disagreed that distance from the transformer was a challenge to electricity connection although non- adopters were the majority. Transformers act as the access points to grid electricity and households within the proximity have a higher probability of adopting electricity compared to those further placed from the transformer. It is noted that majority of non-adopters did not consider the distance from the transformer as a challenge indicating other factors also contributed to non-adoption of electricity. This findings tally with those of Chakrabarti et al. (2002) in West Bengal India, where the cost of connecting electricity to the villages increased with the distance from the grid. During

the household survey one of the respondent noted that although she was 700meters from the transformer, she would have to pay Kshs.150, 000 because her house was outside 600 meters radius. The reason was that her house did not tally with strategy adopted by the rural electrification programme of electrifying customers within 600 meters radius from the transformers, who had to pay Kshs. 35,000.

The study findings revealed that there was a striking dissimilarity across the two categories on the extent that cost of connection was seen as a challenge to electricity adoption. Apparently, 16.7% and 55.56% of adopter and non-adopter households respectively, disagreed that the cost of electricity connection is a challenge compared to 83.3% and 22.2% of adopters and non- adopters‘ respectively, households who saw the cost of electricity connection as a challenge to connectivity. Results further indicated that majority of adopters had no problem with connection while for a good number of non-adopters connection fee was a constraint. Results are in consensus with findings by Barnes (2007) that initial connection charges still remains a challenge when it comes to electricity adoption. More so Karakezi et al. (2008) reported that the most significant barrier to adoption of electricity among the poor is the connection fee.

Findings further revealed that 92.7% and 85.2% of the households‘ disagreed that wiring cost was a challenge to electricity connectivity. Additionally, only 3.7% and

5.2% of surveyed households of adopters and non-adopters did not feel that it was a challenge. These results suggest that wiring cost is not quite a challenge when it comes to electricity adoption.

Findings revealed that 70% and 55.2% of adopters and non-adopters respectively, disagreed that access to information was a challenge in connecting to the grid electricity. The respondents cited that the electricity offices were accessible for any inquiry regarding electricity connection. Another 13% and 24% respectively did not agree. This was due to the fact that some areas were quite remote and sparsely populated and access to information could be difficult.

More so, on paying electricity bills 88.5% and 87% of adopters and non-adopters disagreed that paying electricity bills was a limiting factor in connecting to the grid electricity. Moreover only 11.5% and 13% of adopters and non-adopters respectively, thought that paying bills was a challenge. Hence, households had no problem with paying of electricity bills. This implies that households have no problem in paying their monthly bills as long as they are connected to electricity.

On the other hand, the established the challenges the adopters faced in their day to day use of electricity in their households. The majority (94.4%) of households identified unscheduled power cuts particularly when they lasted for several hours in a day, which affected their use of electricity for household chores and businesses. Power outages lead to destruction of the electronic equipment due to high voltage. This finding concurs with that of Mishra (2010) that households that have electricity connection may have supply of electricity only for a few hours in a day due to inadequate and erratic supply. Other challenges cited by the respondents included delayed reading of metres and overestimation of consumption which was inconveniencing for households.

According to Chaurey et al. (2004) frequent power outages in various areas in rural areas have made households not to discard other sources of lighting even with grid connection, and many resort to other alternative energy sources. This study sought to investigate supplementary sources of lighting that respondent‘s used. The results are presented in (Table 4.14). Findings revealed that most households had multiple sources of lighting.

Table 4.14: Alternative Sources of Energy in the Households

Non-adopters Adopters Response Frequency Percent % Frequency Percent % Kerosene lamp no 6 6.3 22 40.7 yes 90 93.8 32 59.3 Pressurized lamp no 92 95.8 54 100. yes 4 4.2 0 0.0 Candle no 86 89.6 19 35.2 yes 10 10.4 35 64.8 Torch no 3 3.1 0 0.0 yes 93 96.9 54 100.0 Solar no 69 72.6 50 92.6 yes 26 27.4 4 7.4 Source: Field data, 2013

About 100% of adopters owned battery or chargeable torches that they mainly used at night especially when there was need for movement in case of a power outage. A few households had solar panels acquired prior to connecting to the grid electricity.

Kerosene lamps (90%) and torches (93) were used by of non-adopters.

4.5 Effect of Rural Electrification on Public Facilities

The third objective evaluated the perceptions on the quality of service provision in electrified and un-electrified public facilities. Respondents were asked to state whether priority centers (public facilities) in their neighborhood were electrified or not in tandem with rating of general satisfaction in quality of service provision, based on use of various equipment in the facilities. Results showed that in Meru-South Sub-

County, electrification priorities focused on public amenities including; schools, hospitals administrative centers, markets and factories which are seen as focal points and where transformers are to be installed. All the respondents had used the facilities or were familiar with the services provided in these centers. According to the findings, 92.6% of the respondents reported that their nearest market centres were electrified while 7.3% reported that they were not electrified (Table 4.15).

Table 4.15: Responses on Electrified and Un-electrified Public Facilities in Meru-South Sub-County

Percentage of Public facilities Electrified Percent Un- electrified cases Schools 64 42.6% 86 57.3% Health facilities 124 82.6% 26 17.3% Administrative offices 61 40.6% 89 59.3% Market 139 92.6% 11 7.3% Factory 82 54.7% 68 45.3% Total 470 313.1% 280 186.5% Source: Field data, 2013

Additionally, the findings revealed that 82.6% of the respondents reported that the health centers in their neighborhood had electricity while 17.3% reported that they did not have electricity.

About 54.7% of respondents reported that the factories that they used or were close to their households had been electrified while 45.3% reported they were not electrified.

Majority of these were coffee factories and tea collection centres known as kibanda.

Roughly, 42.6% of the respondents reported that schools in their neighborhood were electrified particularly secondary schools while 57.3% reported that schools in their neighborhood were not connected to the grid electricity particularly primary schools.

This was also confirmed by a Kenya Power official in the Sub-County who stated that

72% of learning institutions had been electrified in the region and few primary schools having electricity. It was further reported that a programme designed to connect all public primary schools to the grid was underway. About 40.6% respondents reported that administrative centres had been connected to the grid while

59.3% reported they were not. These facilities included chiefs‘ and sub- chiefs‘ camps, County representative offices, police post and Community Development Fund offices. Additionally, the quality of service provision mean score indicated that electrified facilities had higher ratings of service provision as compared to those of un- electrified facilities (Table 4.16).

Table 4.16: The Mean Scores and Standard Deviation of Quality of Service Provision in Electrified and Un-electrified Public Facilities in Meru-south Sub- County

Status Mean Std Dev Std error mean Schools Electrified 1.94 1.037 0.130 Un- electrified 1.33 0.471 0.051 Hospitals Electrified 4.31 1.149 0.103 Un- electrified 1.69 0.838 0.164 Administration office Electrified 1.72 0.839 0.107 Un- electrified 1.46 0.813 0.086 Markets Electrified 4.04 1.01 0.086 Un- electrified 1.27 0.467 0.014 Factories Electrified 1.87 0.953 1.105 Un- electrified 1.16 0.05 0.409 N=150, (1=poor, 2=fair, 3=Average, 4=good, 5=excellent

Source: Field data, 2013

This implied that majority of the sample respondents using electrified facilities perceived them to be providing better services as compared to the non-electrified facilities.

4.5.1 Quality of Service Provision in Electrified and Un-electrified Public

Facilities

Based on the results, there was significant difference (t-test value=-4.395, p=0.000) in terms of quality of service provision between electrified and un-electrified schools

(Table 4.17).

Table 4.17: Independent Samples t-test for the Differences between Electrified and Un-electrified Public Facilities

Facility t value Sig.(2 tailed) Schools -4.395 0.000* Hospitals -10.993 0.000* Administration office -1.904 0.059 Markets -8.975 0.000* Factories -6.053 0.000* *statistically significant at 5% significance level Source: Field data, 2013

In addition, provision of services in electrified hospitals was significantly different (t- test value=-10.993 at p=0.000) from services delivered in un-electrified hospitals. On the other hand, results indicated that there was no significant difference (t-test value=-

1.904 at p=0.059) in the mean scores of perceived quality of service provision of electrified and un-electrified administrative offices. In relation to market or shopping centers, the perceived quality of service after electrification was equally significantly different (t-test value=-8.975 at p=0.000) compared to when they were un-electrified.

Similar results of significantly different (t-test value= -6.053 at p=0.000) service delivery were observable in the service delivery of electrified and un-electrified factories.

The results indicate that that electrified schools were perceived to have a higher quality of service provision than un-electrified schools. Consequently, the respondents cited various reasons behind their rating scores. The respondents revealed that the major schools that benefited from electrification were secondary schools. For the years that electricity was available in schools, major changes had taken place especially on better and improved lighting. Students studied for longer hours at night with ease and this was perceived to be helping them in attaining higher grades.

Electricity availability also led to the purchase of key equipment for academic use especially computers, used by the students to access information, learn basic computer programmes and also as a curriculum subject. In addition, acquisition of computers by most schools enabled the teaching staff; non-teaching staff (accountants and clerks) improve their service delivery especially on management of school databases. For instance most of the schools that had not been connected to the grid lacked most of such equipment that the electrified schools used hence the difference.

From the results, rural electrification has a direct influence on quality of services in health facilities including dispensaries, health centers as well as hospitals thus improving quality of health care among local communities. In addition, residents attributed electrification of health care facilities to extended time of service delivery which run overnight. In line with these findings, a study by ESMAP (2007) established that nurses in electrified clinics were able to attend to emergencies even at night as the facilities were open for longer hours. Furthermore, respondents‘ positive

perception on improved healthcare delivery could be attributed to enhanced use of electric equipment essential for diagnosis and treatment of complex diseases and illnesses. For instance, medical supplies by health centers could be now refrigerated for preservation which could not previously be done without electricity. Refrigerators in the facilities have facilitated the service of instant immunization of children at health centers of the services and reliability that previously did not have fridges. With reliable source of power, hospitals operated more effectively, more so with the use of sterilizers and electrical detection equipment in rural clinics and more was guaranteed.

During the household interviews a respondent recounts:

‗‗Before the health centre was electrified, i travelled almost 15 km to Chuka district hospital for laboratory tests. But now I am so happy because any time am sick and need to be tested I get tested in Kiereni dispensary which is 1.5km from my home.

(Oral Interviewed, 05/01/2014)’’.

Results on administrative offices imply that quality of service provision in electrified and non-electrified administrative offices is not significantly different. This finding would be expounded on the bases that the respondents reported that they were not at all pleased with the services in administrative offices even after electrification.

Majority of the respondents reported that even after electrification, they still had to print out various security documents especially police abstracts at a higher cost than the usual price. Many of the administration services had not been digitalized even with availability of electricity. Administrative offices are vital in gathering and disseminating information to the rural population, and these study findings raises questions on efficiency of various administrative centers in rural areas following electrification.

Results on market centers indicate that there is a significant difference between electrified and un-electrified markets. Markets are predominant in the area and the majority has diverse activities taking place. Following electrification, a number of stalls started utilizing refrigeration which improved the quality of service delivery

(e.g. storage of fresh drinks) as well as reducing wastage of perishable goods.

To this end, a household head during the interrogation indicated that electricity in their nearest shopping centre has led to opening up of M-pesa and banking agents services which has gradually enhanced transfer of cash; lead to emergence of new services such as; welding, diversified activities such as hair dressing salons, and barber shops; cyber cafes with services such as photocopying, printing and scanning which never existed prior to electrification. There is a clear distinction between electrified and un- electrified markets in service provision where community members benefit from services offered in markets connected to the grid.

The results on factories imply that the respondents were contented with the services delivered in the electrified factories. Meru-South Sub-County is located at the slopes of Mount. Kenya hence it is a rich agricultural area, with major cash crops grown including tea and coffee. Consequently, tea and coffee factories, as well as tea collection points had greatly benefited from provision of electricity. Most of the coffee factories had started processing their own coffee which was not the practice prior installation of electricity where coffee factories sold their coffee to brokers, yielding low profits to the farmers. More so, the tea collection centers had become more secure as delivery could be made at night with reduced security threats.

Consequently, there is thus empirical evidence that electrified public facilities are perceived to be offering higher quality of service provision as compared to un- electrified facilities. However, the observed case of services having not improved could selectively lead to the rejection of the null hypothesis that there is no significant difference in quality of service provision in electrified and un-electrified public facilities as premised by this study. Nonetheless, there is general agreement in the results that the inception of REA initiatives had significantly translated into improved service delivery by most public facilities the services provided by these institutions have also improved as a direct result of the RE.

4.6 Spatial Distribution and Characterization of Transformers, Adopters and Non -

adopters and in Meru-South Sub-County

This section discusses results on the spatial distribution of electricity accessibility

points (transformers), adopters and non-adopters‘ households. Maps have been used

to visualize the distribution of adopters/non-adopters‘ households and transformer

points.

4.6.1 Spatial Distribution and Characterization of Transformers in Meru-

South Sub-County

According to REA (2012) extension of electricity grid to rural areas accompanies

massive installation of transformers, which act as the peak for power distribution

within a specified radius. Results revealed that a total of 35 pole mounted

transformers GPS points were collected from various divisions where the household

survey was conducted. Additionally, all (100%) the transformers are the property of

K.P.L.C who install and maintain them. The results indicated that 89.6% of the

transformers were fully functional by the time of spot check, and 11.4% were non-

operational in the sense that they had been installed but were not functioning as

expected. It emerged that these non-operating transformers had been in this state for

more than a week and the connected households in the area had incurred losses and

inconveniences due to this malfunction and had depended on alternative energy

sources. This erratic power supply discouraged those who had aspirations of adopting

electricity as they cited it as unreliable.

According to the results (Figure 4.11) transformer installation has been spread out

through various years with the 2009 having utmost (37%) transformers installed.

Figure 4.11: Installation of Transformers over Years

35.0%

30.0%

25.0%

20.0%

15.0%

10.0%

5.0% Percentage transformers of Percentage 0.0% 1993 2006 2007 2008 2009 2010 2011 2012 2013 Year of installation

Source: Field data, 2013

In the subsequent years, the decline in transformer installation process declination was mainly associated with inadequate funding for the exercise and also lower demand for electricity connection by the rural households. This drift was well elaborated by the

Kenya Power officials who explained the financial difficulties experienced through this duration. It was until mid 2012 when funds were availed and the process recommenced in 2013 which led to further distribution of electricity and connection in the study area. This study finding concur with that of Haanyika (2008) that efforts in further development of grid electricity infrastructure face a range of challenges such as lack of access to capital.

Further, the results indicated that prior to the year 2009 electricity accessibility was very low with just a few transformers installed near the major roads, major institutions and in the Sub-County headquarter offices. The spatial distribution revealed by using

a map (Figure 4.12) exemplified disparity in the spatial distribution of transformer points.

Figure 4.12: Spatial Distribution of Transformers in Meru-South Sub-County

Source: Author, 2013

Evidently, the west and south western part of the Sub -County appeared to be well served with the transformers as compared to the other parts. Majority of transformers were found to be linearly distributed, of which majority had been mounted along main rural roads while a few penetrated to the interior. It was also established that a good number of the transformers were located in priority centers such as markets, towns, schools and hospitals, especially those in the western and central parts of Chuka and

Magumoni division. This concurs with findings by Abdullah and Jeanty (2011) that

most grid electricity transformers and distribution lines in rural areas are located in commercial or major trading areas, alongside tarmacked public roads.

Various factors were attributed to the observed spatial distribution of transformers in the study area. First and foremost, the western region lies in the Upper Midland Agro- ecological zones (Jaetzold et al., 2006) on the eastern slopes of Mount Kenya and borders the Mount Kenya forest and also densely populated (KNBS, 2009). This is a geographical advantage which makes the area more likely to have higher accessibility rates because of the cost effectiveness in distribution of electricity via grid extension in such a region. According to Parshall et al. (2009) a higher population density is attractive when it comes to grid electrification as it reduces inter-household distance hence; lower amount of low voltage line is needed to connect to each household.

Therefore, the average connection cost is reduced for lower inter-household distances.

Additionally, higher accessibility near towns could have been attributed to the proximity of these regions to the major towns usually the growth points and areas in the vicinity tend to have a better opportunity for a range of infrastructural development such as electricity (Pellegrini and Tasciotti, 2013).

The eastern part of the study area also exhibits a distinctive pattern where there are fewer transformers in the study area. This area is arid/semi arid and is sparsely populated with low population density. In essence, this may be a factor contributing to the installation of fewer transformers since a low population density acts as a barrier to grid extension because of the expenses associated with vast distance (Ahlborg and

Hammar, 2011). Likewise Pellegrini and Tasciotti (2013) explains that low population density is a barrier in increasing access to electricity in rural areas

because of the high capital and operating costs for electricity companies and very low return to investments. More so, households in such areas are often poor with low electricity consumption hence electricity accessibility may not be prioritized for such regions (Pellegrini2013). Electricity distribution cost is spread over relatively few people resulting in high expenses for each unit of electricity consumed. Generally, electricity access level need to be increased as total 100% accessibility has not been achieved as over the last five years as it was indicated in the rural electrification strategic plan (REA, 2013). The implication of these results is that there is need to review and adopt a better rural electrification plan, which includes aspects such as settlement patterns which would otherwise facilitate homogeneous electricity accessibility in the area.

4.6.2 Spatial Distribution of Adopters in Meru-South Sub-County

The result showed distinctive disparity in the spatial distribution of adopters‘ households in the study area whereby, adopters are mainly in the west and south western region, with very few stretching out in the South eastern parts Meru-South

Sub-County (Figure 4.13). A linear distribution is exhibited where the majority of the adopters are clustering along the major roads, shopping centers and health centers.

Figure 4.13: Spatial Distribution of Adopters in Meru-South Sub-County

Source: Author, 2013

More so, adopters‘ households seem to be clustering where there are transformers hence this scenario can be expounded on the basis that households near electricity supply points have got an added advantage due to the advantage of proximity.

Adoption is much lower in the lower zones because even in a few areas that have been mounted with full functional transformers there were absolutely no sampled adopters.

4.6.3 Spatial Distribution of Non-Adopters in Meru-South Sub-County

According to the survey results, a high number of households are not connected to the grid electricity (64%) (Figure 4.14).

Figure 4.14: Spatial Distribution of Non-Adopters in Meru-South Sub-County

Source: Author, 2013

Spatially both the upper and lower zones exhibit a high stretch of non-adoption. This pattern is unique because even in areas with high accessibility, some of the households are not connected to the grid. This indicates that despite households having access to electricity other vital factors influence electricity adoption (Louw et 80

al., 2008). In essence, Barnes (2007) elaborates that households in electrified villages may not be able to connect, even though the connection is available for more than 20 years, due to high connection fee. Generally, majority of households are not connected to the grid electricity despite their proximity to electricity access points.

CHAPTER FIVE

SUMMARY OF FINDINGS, CONCLUSIONS AND RECOMMENDATIONS

5.1 Introduction

This chapter summarizes the study findings and makes conclusions drawn from them.

Recommendations and areas for further research are also presented. The summary of the findings is in tandem with the objectives of the study.

5.2 Summary of Key Findings

The study surveyed a total of 150 households in Meru-South Sub-County of which

36% and 64% were of electricity adopters and non-adopters respectively. Meanwhile, the study considered the following household socio-economic characteristics influences electricity adoption using binary logistic regression model: distance from the transformer, gender, age, marital status, education, occupation, family size, house type, source of income, and monthly income. The results indicated that distance from the transformer, household size, gender, education, and total monthly income as the predictor variables for electricity adoption.

On assessing the socio-economic benefits of electricity adoption among households results revealed that all (100%) the adopter households had acquired lighting devices.

Mobile phones, radio and television ownership stood at 98.1%. 96.3% 94.4% respectively. The least owned electric appliances were those used to ease domestic labour which included electric iron box, refrigerator, chaff cutter, electric heater and computer ownership at 27.8%, 9.3% and 27.8% 11.1% and 3.7% respectively.

Concerning the direct socio-economic benefits of electricity, the study revealed that the benefits which brought about improvement in quality of life was the most prevailing among households and those that come about due to the productive uses of electricity were fewer especially in agriculture and cottage industry sector. For instance all connected households with lighting devices used them in lighting spaces in the household, and this was the first priority in electricity use following connection.

Apart from indoor illumination, about 27.8% of the households used electricity for security lighting.

Radios were used in accessing information. Moreover 7.4% also used radios for entertainment purpose. Ownership of mobile phone was also established to have improved communication. Television set was also a common electric appliance among households whose major use was entertainment as reported by 81.5% of households and access to information at 79.6%.

Several other electrical appliances owned by a smaller proportion of households included refrigerator (9.3 %), computers (3.7%), chaff cutter (27.8%) and electric heater (11%).

Household members were involved in several small businesses that used electricity connected to the household. The study results revealed that a few households (18%) of households had home businesses among those affirmed having business at home about 77.8% (21) were adopters and 22.2% (6) are non-adopters. The mobile phone charging business was the most prevalent at 44.74%, hair dressing and barber shops each at 10.53%, and 34.21% of households respectively had general shop businesses.

Home-based businesses provided income for the households. The salon business had an average monthly total income of Kshs. 6,000 barber shop business mean monthly income Kshs. 5,250. Mobile phone business being the most prevalent was run by 17 households and the mean monthly income was 8,500. Ownership of general shop was the second most common home business with 13 households running it and having a monthly income of 14,769 hence having the highest mean monthly income among all small businesses conducted by households. Profit got was utilized in various ways.

Paying school fees for the children was the primary expenditure item was at 86%.

Paying electricity bills was at 36.4% while for domestic use was at 63.6%.

On assessing the challenges faced in electricity adoption, the respondents‘ perception on how various aspects challenge electricity adoption was assessed using six statements on a 5-point Likert scale: +5 (strongly agree) to +1 (strongly disagree).

The study revealed that adopters and non-adopters had a strong conviction with the fact that distance from the transformer distance is a challenge when it comes to electricity connection from the grid to their households. This point out that majority up hold that distance from the transformer really challenge to electricity connection.

On the aspect of the cost of connection also both adopters (85.4%) and non-adopters

(88.9%) strongly agree that the cost of electricity connection is a challenge to electricity adoption. This indicates that connection cost still remains a challenge when it comes to adoption. It was more so revealed that 92.7% and 85.2% of adopters and non-adopters reported to disagree with aspect of wiring cost being a challenge to electricity connectivity in their households. These results imply wiring cost is not quite a challenge when it comes to electricity adoption. Majority of households in

rural areas are always ready to do the wiring in their houses as long as electricity is available.

On the aspect of delayed electricity installation as a challenge to electricity connection, about 89.65% and 83.3% respectively of the adopter and non-adopter household strongly disagreed and precisely 3.7% and 2.1% of adopter and non- adopter respectively, strongly agreed with this aspect. This indicates that households that applied for connection were connected at the right time range. This suggests that

Kenya Power is very effective in connecting households that have settled all their dues.

The study revealed the extent which access to information on connection procedure challenged connection to the grid electricity. The study results reveal that 70.0% and

55.2% of adopters and non-adopters respectively disagreed that access to information was a challenge in connecting to the grid electricity. The respondents indicated that the electricity offices were accessible for any inquiry regarding electricity connection.

Another 13% and 24% of adopters and non-adopters respectively strongly agreed that access to information is a challenge to electricity connection to the households.

On the payment of bills, paying bills 88.5% and 87% of sampled households of adopters and non-adopters sampled respectively, disagreed with the contention that paying bills was a limiting factor in connecting to the grid electricity. Moreover

11.5% and 13% of adopters and non-adopters strongly disagreed that paying bill was a challenge when it came to connection of electricity to their households. It is noteworthy that electrified and non-electrified households have similar perception on

this crucial factor. This suggests that households have no problem in paying their bills as long as they are connected to electricity. The respondents reported to have experienced unscheduled power cuts where electricity was cut for a whole day without any notice. Other challenges sighted by the respondents were delayed reading of bill and overestimation, power siege and most of the electronic equipment were destroyed due to high voltage and power outage.

In the assessment of the effect of rural electrification on public facilities, the results showed that the priority public amenities that had been electrified included; schools hospitals administrative centers markets and factories. About 92.6%, 82.6%, 54.7%,

42.6% and 40.6% of the respondents respectively reported that their nearest market centers, health centres, factories, schools and administrative centers were electrified.

In establishing the difference in quality of service provision in electrified and un- electrified public facilities, data was subjected to further analysis by use of independent sample t test at 95% confidence level. These results further indicated majority of the respondent using electrified facilities perceived them to be providing better services as compared to the non-electrified facilities. There was a significant difference between electrified and non-electrified schools, markets, health facilities, offices and factories whereas, no significant difference was revealed in the case of administration offices.

On assessing the spatial distribution and characterization of electricity adoption, non- adoption and accessibility in the study area, a total of 35 pole mounted transformers

GPS points were recorded from the various divisions where the random household survey was conducted. 89.6% of the transformers were fully functional by the time of

spot check, and 11.4% were non-operational in the sense that they had been installed but were not functioning as expected. A map generated on the spatial distribution of transformer points show that the west and south western parts of the Sub-County appeared to be well served with the transformers as compared to the other parts. The transformers were lineally distributed. Additionally, it was established that a good number of transformers are located near priority centers such as markets, towns, schools and hospitals, especially those in the western and central parts of Chuka and

Magumoni division.

A map generated on the spatial distribution of electricity adopters, shows distinctive disparity in the spatial distribution of adopter households in the study area. Majority of adopters are in the upper and middle zones with very few stretching out in the lower zones. A linear distribution is exhibited where the majority of the adopters tend to cluster along the major roads, shopping centers and health centers. More so, adopter households seem to be concentrated around the transformers.

According to the survey results, a high percentage of households are not connected to the grid electricity. Spatially, both the high and lower zones display high percentage of non-adoption. This pattern is unique because even in areas with high accessibility a good number of households are not connected to the grid.

5.3 Conclusion

Results presented in this study have established that there is generally low electricity adoption and that households‘ socio-economic characteristics such as household size, gender of household head, education attainment and distance from the transformer

were the key determinants of electricity adoption in Meru-South Sub-County as explained by binary regression model. With these results the null hypothesis that there is no significant relationship between households‘ socio-economic characteristics and electricity adoption was rejected. The results indicated that rural electrification adoption had brought substantial benefits to the households; many households have improved quality of life especially from lighting which was practiced by all the adopters‘ households. It was also established that few households owned home business that utilized electricity and reported to be of high benefit to their families.

Major concerns observed were challenges that the households experienced which included blackouts, high connection cost and the distance from the electricity access points. Respondents both adopters and non-adopters acknowledged that the aforementioned factors discriminated electricity adoption, indicating that these challenges were of great concern. From the empirical findings it is evident that public facilities connected to grid power are perceived to have higher quality of service provision as compared to those facilities not connected to electricity. This leads to the rejection of the null hypothesis that there is no significant difference in quality of service provision in electrified and non-electrified public facilities.

The results further revealed that there was a remarkable spatial heterogeneity in access adoption and non-adoption patterns within rural households. The upper land regions have a higher accessibility compared to the lower regions. Additionally, some areas having a higher share of households without access to electricity than others.

Consequently, even in areas with higher electricity access level large shares of the rural households do not use electricity. The spatial information integrated with the descriptive analysis has assisted in visualizing and comprehending the adoption, non-

adoption and accessibility distribution patterns in the region. Conclusively, socio- economic characteristics of households are key aspects in determining adoption of grid extended electricity.

5.4 Recommendations

From the study findings the following recommendations are made:

· There is need to ease the connection charges by for example subsiding the cost of electricity connection or by providing long term fee spread over years.

· Rural electrification planning process should involve assessment of the potential for productive uses of electricity for households and social services and include measures for their promotion during. Promotion of and capacity building for productive uses of electricity in rural areas can increase the productivity of rural businesses, enable a more efficient use of the supply infrastructure, and improve the revenues of distribution companies, thereby enhancing the economics of electrification.

· The cost and availability of electric appliances, such as cooking stoves, has often been a prohibiting factor in the uptake of electricity. It is then recommended that appliance costs should be subsidized and be locally available to increase the demand for electricity among households.

5.5 Areas of Further Research

The following areas are recommended for further research:

· There is need to assess and identify types and the distribution patterns of public infrastructural facilities in Meru-South Sub-County and also to determine their influence on electricity distribution.

· There is need to evaluate the design and the planning for rural electrification distribution in the study area using GIS techniques.

REFERENCES

Abdullah, S., & Jeanty, P. W. (2011). Willingness to pay for renewable energy: Evidence from a contingent valuation survey in Kenya. Renewable and Sustainable Energy Reviews, 15(6), 2974-2983.

Abdullah, S., & Markandya, A. (2012). Rural electrification programmes in Kenya: Policy conclusions from a valuation study. Energy for Sustainable Development, 16(1), 103-110.

Ahlborg, H., & Hammar, L. (2011).Drivers and barriers to rural electrification in Tanzania and Mozambique–grid extension, off-grid and renewable energy sources. In World Renewable Energy Congress. Linköping (p. 8). Annual Review of Environment and Resources 32: 169 203.

Ambunda, L., & de Klerk, S. (2008). Women and custom in Namibia: A research overview. na.

Andreas, K. (2006). Regional disparities in electrification of India–do geographic factors matter?. Center for Energy Policy and Economics, Working Paper, (51).

Anríquez, G., &Stamoulis, K. (2007). Rural development and poverty reduction: Is agriculture still the key?Available.at:http://www.the-star.co.ke/news/article- 17279/rea-says-electricity access-73cent#sthash.9NrlTmCS.dpufAugust 25, 2013).

Badiani, R., & Jessoe, K. K. (2011). Electricity subsidies for agriculture: Evaluating the impact and persistence of these subsidies in India DRAFT.

Barkat A.( 2003). Rural electrification and poverty reduction: case of Bangladesh. Presented at NRECA Int. Conf., Sustain. Rural Electrification in Developing Countries: Is it Possible, Arlington, VA.

Barnes, D. F. (Ed.). (2007). The challenge of rural electrification: strategies for developing countries. Washington DC, Earth Scan.

Barnes, D., (2005). Transformative Power: Meeting the Challenge of Rural Electrification. ESMAP Knowledge Exchange Series No. 2.The World Bank. Washington, DC.

Barnes, D.K., Krutilla, and W.F. Hyde (2005). The Urban Household Energy Transition: Social And Environmental Impacts in the Developing World. Washington D:C: Resources for the Future Press.

Barrios, E. B. (2008). Infrastructure and rural development: household perceptions on rural development. Progress in Planning, 70 (1), 1-44.

Bekker, B., Eberhard, A., Gaunt, T., & Marquand, A. (2008). South Africa's rapid electrification programme: Policy, institutional, planning, financing and technical innovations. Energy Policy, 36(8), 3125-3137.

Bernard, T., Torero, M. (2009). Impact of Rural Electrification on Poorer Households in Ethiopia. Washington DC, Report for the World Bank.

Bernard, T. (2012). Impact analysis of rural electrification projects in sub-Saharan Africa. The World Bank Research Observer, 27(1), 33-51.

Bhattacharyya, S. C. (2006). Energy access problem of the poor in India: Is rural electrification a remedy?. Energy Policy, 34(18), 3387-3397.

Bhattacharyya (Ed.). (2013). Rural Electrification through Decentralized Off-grid Systems in Developing countries, Green Energy and Technology. Verlag, London: DOI: 10.1007/978-1-4471-4673-5_2, _ Springer- London.

Bose, T. K., Uddin, M. R., & Mondal, A. (2013). Impacts of Electricity Access to Rural SMEs. International Journal of Managing Value & Supply Chains, 4 (4).

Cabraal, R. A., Barnes, D. F., & Agarwal, S. G. (2005).Productive uses of energy for rural development.Annu. Rev. Environ. Resour.,30, 117-144.

Cecelski, E., & Ounalli, A. (2006). Low-cost Electricity and Multisector Development in Rural Tunisia.Sub-Saharan Africa: Introducing Low-cost Methods in Electricity Distribution Networks, 183.

Christensen J., Gordon M.,and Mathilde B. (2012). Enhancing access to electricity for clean and efficient energy services in Africa. Nairobi, UNEP.

Chakrabarti, S., & Chakrabarti, S. (2002). Rural electrification programme with solar energy in remote region–a case study in an island. Energy Policy, 30 (1), 33-42.

Chaurey, A., Ranganathan, M., & Mohanty, P. (2004). Electricity access for geographically disadvantaged rural communities—technology and policy insights. Energy policy, 32(15), 1693-1705.

Clancy, J. (1999). Policies, projects and the market empowering women. Some initial reactions to development in the energy sector.

Clancy, J. (2003). Gender and household energy concerns: the global context. Briefing paper 2 for SPARKNET.

Creative Research Systems, CRS (2009). Sample Size Formula for Sample Size Calculator. Available at: http://www.surveysystem.com/sscalc.htm. Accessed on: 05/10/2013.

Davis, M. (1998). Rural household energy consumption: the effects of access to electricity Evidence from South Africa. Energy Policy, 26(3), 207-217.

Dreze, J. and P.V. Srinivasan (1997), ―Widowhood and Poverty in Rural India: Some Inferences from Household Survey Data‖, Journal of Development Economics, 54(2), 217-234.

Dungumaro, E.W. (2008). Gender differentials in household structure and socioeconomic characteristics in South Africa. Journal of Comparative Family Studies, 39(4):429-451.

Elias, R. J., & Victor, D. G. (2005). Energy transitions in developing countries: a review of concepts and literature. Program on Energy and Sustainable Development, working paper. Stanford University: Stanford.

Enrique, E. (2010). Addressing the electricity gap: World Bank Energy Sector group.Retrieved on May 01, 2013.

ESMAP. (2007). Technical and Economic Assessment of Off‐grid, Mini‐grid and Grid Electrification Technologies.Technical Paper. Washington, DC: Energy Sector Management Assistance Program, The World Bank.

Fishbein, R. E., Sanghvi, A. P., & Unit, A. E. (2003). Survey of productive uses of electricity in rural areas.Washington, DC.

Fuwa, N. (2000). The poverty and heterogeneity among female-headed households revisited: the case of Panama. World Development, 28(8), 1515-1542.

Gibson, J., & McKenzie, D. (2007).Using global positioning systems in household surveys for better economics and better policy.The World Bank Research Observer, 22(2), 217-241.

GoK (Government of Kenya), (2014). Economic Survey 2013. Nairobi, Government printer.

Hair, J., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Upper Saddle River, New Jersey: Pearson Education International.

Haanyika, C. M. (2008). Rural electrification in Zambia: A policy and institutional analysis. Energy policy, 36(3), 1044-1058.

Heltberg, R. (2003). Household fuel and energy use in developing countries: A multi- country study. World Bank Oil and Gas Policy Division. Washington, DC: World Bank.

Heltberg, R. (2005). Factors determining household fuel choice in Guatemala. Environment and Development Economics, 10(3), 337-361.

Hemmings, A. (2006). Great ethical divides: Bridging the gap between institutional review boards and researchers. Educational Researcher, 35(4), 12-18.

Hilbe, J. M. (2009). Logistic regression models. Boca Raton: CRC Press.

Hing, V., Saing C., & Khieng S. (2012).Baseline Survey for Socio-economic Impact Assessment: Greater Mekong Sub-region Transmission Project CDRI Working Paper Series, 70).

Horgan, G. (2007). The impact of poverty on young children's experience of school. Joseph Rowntree Foundation.

Independent Evaluation Group (2008). The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits. An Impact Evaluation. World Bank,Washington, D.C., Independent Evaluation Group (IEG).

International Energy Agency (2012), World Energy Outlook 2012 – Measuring progress towards energy for all: power to the people. Paris, International Energy Agency. International Energy Agency (2011), Energy for All, Financing Access for the poor, Special Excerpt of the World Energy Outlook 2011, International Energy Agency. Paris.

International Energy Agency.(2010). Electricity Access in 2009-- Regional Aggregates. Retrieved from IEA: http://www.iea.org/weo/electricity.asp Accessed 14 July 2013).

Jan, I. (2012). What makes people adopt improved cookstoves? Empirical evidence from rural northwest Pakistan.Renewable and Sustainable Energy Reviews, 16 (5), 3200-3205.

Jaetzold, R., Schmidt, H., Hornet, Z.B., Shisanya, C.A. (2006). Farm management handbook of Kenya. Natural conditions and farm information, vol 11/C, 2nd edn. Ministry of agriculture/GTZ, Nairobi (Eastern Province)

Johansson, T., José G, (2004). World Energy Assessment: Overview 2004 Update, UNDP/UN-DESA/World Energy Council, 500 pages—United Nations Development Programme, New York.

Karekezi, S., Kithyoma, W., & Initiative, E. (2003). Renewable energy development. In workshop on African Energy Experts on Operationalizing the NEPAD Energy Initiative, June (pp. 2-4).

Karekezi, S., Kimani, J., & Onguru, O. (2008). Energy access among the urban poor in Kenya. Energy for Sustainable Development, 12(4), 38-48.

Katuwal, H., & Bohara, A. K. (2009). Biogas: a promising renewable technology and its impact onrural households in Nepal. Renewable and Sustainable Energy Reviews, 13(9), 2668-2674.

Kemausuor, F., Brew-Hammond, A., Obeng, G. Y., Duker, A., Annor, F. O., Boamah, F., &Ladzagla, D. (2012).GIS-based Support for Implementing Policies and Plans to Increase Accessto Energy Services in Ghana.

Kembo, V.S. (2013) Socio-economic Effects of Rural Electrification In Tala Division, Machakos County, Kenya. Unpublished MA thesis: University of Nairobi.

Kemmler, A. (2007). Factors influencing household access to electricity in India. Energy for Sustainable Development, 11(4), 13-20.

Kenya National Bureau of Statistic. (2009). Kenya Household Survey and Population Census Nairobi: Government Printer.

Ketlogetswe C., Mothudi T.H., Mothibi J. Effectiveness of Botswana's Policy on Rural Electrification. Energy Policy (2007); 35(2):1330-7.

Khandker, S. R., Barnes, D. F., Minh, N. H., &Samad, H. (2009).Welfare impacts of rural electrification: evidence from Vietnam. Washington DC, World Bank.

Kooijman-van Dijk, A. L., & Clancy, J. (2010). Impacts of electricity access to rural enterprises in Bolivia, Tanzania and Vietnam. Energy for Sustainable Development, 14(1), 14-21.

Kopp, A., & MAS-cycle, N. A. D. E. L. (2010). How to use rural electricity in a productive way. NADEL MAS-Cycle 2008–2010.

Kowsari, R., & Zerriffi, H. (2011). Three dimensional energy profile: A conceptual framework for assessing household energy use. Energy Policy, 39(12), 7505-7517.

Leech, N.L., Barrett, K.C., & Morgan, G.A. (2005). SPSS for Intermediate Statistics: Use and Interpretation (2 nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Publishers.

Louw, K., Conradie, B., Howells, M., & Dekenah, M. (2008). Determinants of electricity demand for newly electrified low-income African households. Energy Policy, 36(8), 2812-2818.

Mbatia J. K., (2011). The socio-economic effects of the rural electrification programme: a case study of Kandara location in Maragua District, Central Kenya. Unpublished MA Thesis: University of Nairobi.

Maleko, G. C. (2005). Impact of electricity services on micro-enterprise in rural areas in Tanzania. Unpublished MA thesis:University of Twente, Enschede.

Mapako, M., & Prasad, G. (2007). Rural electrification in Zimbabwe reduces poverty by targeting income-generating activities.

Mekonnen, A., & Köhlin, G. (2009). Biomass fuel consumption and dung use as manure: Evidence from rural households in the Amhara region of Ethiopia. Institutionen för nationalekonomi med statistik, Handelshögskolan vid Göteborgs universitet.

Mills, B. and J. Schleich (2010), Why Don‘t Households See the Light? Explaining the Diffusion of Compact Fluorescent Lamps, Resource and Energy Economics 32(3), 363-378.

Mishra, P. K. (2010). Alleviating Energy Poverty Through Innovation: The case of Jyotigramyojana (Rural lighting Scheme) of Gujarat.World Energy Congress.

Moghadam, V. M. (1998). Feminisms and development.Gender& History, 10(3), 590- 597.

Mvondo, J. M. (2010). Impact of access to free basic electricity on household‘s poverty in Buffalo City Municipality in the Eastern Cape. Unpublished Doctoral dissertation, University of Fort Hare.

Niez, A. (2010). Comparative Study on Rural Electrification Policies in Emerging Economies: Keys to Successful Policies (No. 2010/3). OECD Publishing.

Nishimwe-Niyimbanira, R., Sekhampu, T. J., & Paul-Francois, M. (2014). A Gender Comparison of Access to Basic Necessities in a South African Township: Application of the Lived Poverty Index. Mediterranean Journal of Social Sciences, 5(8), 94.

Ogalo, K. (2011) Factors Influencing Electricity Distribution in Nyamarambe Division, Kisii County, Kenya. Unpublished MA thesis: University of Nairobi.

Pachauri, S. (Ed.). (2007). An energy analysis of household consumption: Changing patterns of direct and indirect use in India (Vol. 13). Springer.

Park, H. M. (2009). Comparing group means: t-tests and one-way ANOVA using Stata, SAS, R, and SPSS. Technical Working Paper. The University Information Technology Services (UITS) Centre for Statistical and Mathematical Computing, Indiana University.

Parshall, L., Pillai, D., Mohan, S., Sanoh, A., & Modi, V. (2009). National electricity planning in settings with low pre-existing grid coverage: Development of a spatial model and case study of Kenya. Energy Policy, 37(6), 2395-2410.

Pellegrini, L., & Tasciotti, L. (2013, March). Rural Electrification Now and Then: Comparing Contemporary Challenges in Developing Countries to the USA's Experience in Retrospect. Forum for Development Studies (Vol. 40, No. 1, pp. 153- 176). Routledge.

Peng, C. Y. J., Lee, K. L., & Ingersoll, G. M. (2002). An introduction to logistic regression analysis and reporting. The Journal of Educational Research, 96(1), 3-14.

Peters, J., Harsdorff, M., & Ziegler, F. (2009). Rural electrification: Accelerating impacts with complementary poor. Paris, International Energy Agency.

Rural Electrification Authority, (2013).Retrieved on 22 February 2013 from http://www.rea.com/.

Schlag, N., & Zuzarte, F. (2008). Market barriers to clean cooking fuels in Sub- Saharan Africa: a review of literature. Stockholm Environment Institute, Stockholm.

StatSA. 2010. Social profile of South Africa, 2002–2009. Pretoria: Government Printer.

Thom, C. (2000). Use of grid electricity by rural households in South Africa. Energy for Sustainable Development, 4(4), 36-43.

Victor, D.,(2009). The Politics of Fossil Fuel Subsidies. GSI, Untold Billions: Fossil- fuel subsidies, their impacts, and the path to reform. Geneva, October 2013. http://www.globalsubsidies.org/files/assets/politics_ffs.pd

Wang, F., & Luo, W. (2005). Assessing spatial and non spatial factors for healthcare access: towards an integrated approach to defining health professional shortage areas. Health & place, 11(2), 131-146.

Wanyoike, M. (2012). Factors influencing electricity connectivity in rural Kenya: a case of Mt. Kenya south sub-region of the Kenya power. Unpublished Doctoral Dissertation, Nairobi, University of Nairobi.

Wamukonya, N., & Davis, M. (2001). Socio-economic impacts of rural electrification in Namibia: comparisons between grid, solar and un-electrified households. Energy for Sustainable Development, 5(3), 5-13

Whitfield& Darby (2006) Household Behavior and Energy Demand: Evidence from Peru, in Public Policy Massachusetts, Harvard (2006) p. 143.

Winkler, H., Simões, A. F., Rovere, E. L. L., Alam, M., Rahman, A., &Mwakasonda, S. (2011). Access and affordability of electricity in developing countries. World Development, 39(6), 1037-1050.

World Bank, (2010).Trading Economics: Access to electricity in Kenya. WB Aid for Energy Sector.

Ying, W. Z., Hu, G., &Dadi, Z. (2006). China's achievements in expanding electricity access for the poor. Energy for Sustainable Development, 10(3), 5-16.

Zibran, M. F. (2012). Chi-Squared test of independence. Department of Computer Science, University of Calgary, Alberta, Canada.[online].[Cited 2013-08-12].

Zhang, X., & Kumar, A. (2011). Evaluating renewable energy-based rural electrification program in western China: Emerging problems and possible scenarios. Renewable and sustainable energy reviews, 15(1), 773-779.

APPENDICES:

APPENDIX I: SAMPLE QUESTIONNAIRE

Hello, I am a Master of Art (Urban and Regional planning) student at Kenyatta University, Nairobi. As part of my MA research thesis requirements, I am conducting a survey on rural electrification adoption determinants. The information you provide will be treated with utmost CONFIDENTIALITY. Your assistance in answering these questions truthfully will be highly appreciated. Thank you

Enumerator’s Name: ______Date of interview____/_____/ 2013 Starting time______:______Household number______GPS coordinates: S: ______o.______‘.______‖ E: ______o.______‘.______‖ Altitude (meters above sea level) ______

No. Variable label Variables Skip rules, Information, remarks Identifying variables 1. Division 1=Chuka 2=Magumoni 3=Igamba Ng’ombe 2 Location ………………………………. 3. Sub-Location and village ………………………………. 4. Name of household Head (main decision maker) 5. Mobile phone number

Section one A: Household socio economic and demographic characteristics 6. Gender of the 1=Male household head 2=Female

7. Relationship of the 1=Household head respondent to 2=Spouse household head 3=Own child 4=Relative5=Other(specify) 8. Age of the household head (years 9. Marital status 1=Single,2=Married,3=Single parent,4=widowed,5=separated 6=Other (specify)

10. Educational levels of 1=None, 2=primary education, the Household head 3=secondaryeducation,4=⁮College/Diploma 5= 1st Degree,6= 2nd and PhD Degree 6= Other (Specify) 11. Main occupation of 1=Employed, 2=Self –employed,3=Daily the household head laborer, 4=Unemployed/pensioner,5=Other (Specify) 12. No. of members of the household (family size) 13. Main type of dwelling 1 =Bricks,2=Stone,3=Earth/Mud,4=Wooden unit walling material 5 = Iron sheet 6=Cement blocks,7=Others, Specify…………………….. 14. Main sources of income 15. Monthly total income 1=Below 5000 (Kshs 2=5.001 - 10,000 3=10,001 - 20,000 4= 20,001 - 30,000 5 =Over 30000 Section One B: Household electricity adoption/Non-Adoption 16. Is your household 1=Yes, 2=No connected with grid extended electricity? 17. Why are you not currently connected to electricity? 18. Why did you connect 1=better services,2=Lower to grid electricity monthly expenditure,3=Others 19. For how long has the household been connected to electricity? (years)

20. How did you get 1=Community leaders, 2=RE information on the Officials, 3 =Community connection members, 4=Media, 5=Others procedure? (specify…………………. 21. Where did you get 1=Savings 2=Loan,3=Funded the funds in aid of 4= Others(specify) adopting electricity? 22. How much do you 1=Less than 500 pay for your 2=500 to 1000 household electricity 3= 1000 and above consumption on a monthly basis? (Kshs.)

23. Which initiative are 1=Umeme Pamoja, 2=Stima Loan, you aware of that is 3=All, 4=None, 5=Others promoting (specify)…………………… rural electrification adoption in the study area? 24. Have you 0=No participated in any of 1=Yes the initiatives? 25. Distance from the nearest transformer (km) Section Two A: Socio economic benefits of electricity 26. Which household electric appliances do you own/and No. of times used/Benefits? Appliance (0=No;1=Yes) Frequency Benefits of Codes A services a) Radio set b) TV/DVD set c) Refrigerator d) Washing machine e) Electric iron f) Mobile phones g) Cooking/ heating /cooling devices h) Computers/printers i) Electric water pump j) Lighting devices (bulbs, lamp shades) k) Motorized agricultural machinery l) Others (Specify)………………….. Codes A 1=Daily2=Weekly 3=Monthly4=Quarterl y 5=Yearly 6= Never 27. Do you have a business at home? 0=No,1=Yes

28 If yes what type, the no of hours worked every day and total monthly income

Services No of hours worked every Total monthly income Increase in profit Code A week 1=Yes;0=No

Codes A 1= Hair salon, 2= Hair cut for men,3= Charging phone, 4= Tailoring,5= Crop processing, 6=Electronics repair, 7= Shop owner, 8=Others(specify)

100

29. How is the income got utilized……………………………………………………

30. Rank the benefits of Strongly Agree Don’t Disagree Strongly electricity agree know disagree 1) Electricity in household is good for children education

31. Facilities 0=Non- Have you ever used or familiar with services this public electrified, facility? 1=yes,0=no .If yes how would you rate the 1=electrified overall satisfaction on quality of the service provided in the public facility based on equipment/appliances used? If no skip to question 32) (circle appropriately) 1) Schools 1………..2………..3….……4……..5 Explain your rating ………………………………………………………………………………………………… …… 2) Hospitals 1………..2………..3….……4……..5 Explain your rating ………………………………………………………………………………………………… … 3) Administr 1………..2………..3….……4……..5 ation offices Explain your rating ………………………………………………………………………………………………… ….. 4) Markets 1………..2………..3….……4……..5 Explain your rating ………………………………………………………………………………………………… …… 5) Factories 1………..2………..3….……4……..5

Explain your rating ………………………………………………………………………………………………… …… Section Two B: Challenges to electricity adoption 32. Have you ever been disconnected for not paying your bills? 0= No,1=Yes 33. How many days have you experienced power outage in the last 0= No,1=Yes one week?(blackout) 34. How long does the power outage last? 1=Never,2=Once,3= Two times,4=Three or more time 35. What are some of the challenges that you have experienced in adopting electricity and rate .(explain) Distance from the transformer Cost of connection Wiring cost Installation Information

101

Paying bills Others(specify)………………………………………………… …………. 36. Do you experience unscheduled power cuts. 0=No,1=Yes 37. Approximately how long do they last? 1=< 12hrs,2=24hrs,3= 2-4 days,4=above 5 days,5=one week

38. What other sources do you use for lighting? 1. Kerosene/diesel lamp 0=No Tick all that apply 1=Yes [√] 2. Pressurized lamp 0=No 1=Yes 3. Candles 0=No 1=Yes 4. Torches 0=No 1=Yes 5. Others (Specify)

Time the interview ended………………………….. Thank you for your cooperation!

102

APPENDIX II: IN-DEPTH INTERVIEW GUIDE INVOLVING ELECTRICITY AGENCIES /LOCAL ADMINISTRATORS IN MERU - SOUTH SUB-COUNTY PREAMBLE

Name of the interviewee: ______Date of interview____/_____/ 2013 Starting time ………………

EFFECT OF RURAL ELECTRIFICATION ON SERVICE DELIVERY

INPUBLIC FACILITIES, THE PROSPECTS AND CHALLENGES IN THE

STUDY AREA

1. What are some of the facilities that have been established since rural electrification was rolled out in the study area?

2. What are some of the facilities that have been have been electrified in the study area?

3. Have all the targeted public facilities been electrified in the study area?

4. What are some of plans in promoting adoption of rural electrification?

5. Has there been any progress towards rural developments following rural electrification?

6. What is the local administration participation in promoting rural electrification in the study area?

7. Has the RE policies and initiatives had any implication on rural development?

8. Have the residents participated fully in productive uses of electricity in the study area?

Thank you for taking time to answer these questions to have this study complete successfully. Ending time……..… ………

103

APPENDIX III: TRANSFORMER ATTRIBUTE SHEET

Date......

Division...... Location...... Sub- location...... Point No......

GPS coordinates: S: ......

E: ......

Altitude......

A) Transformer point characteristics

Year of installation......

Current operational status: a) Fully Operational b) Non-operational (Tick where applicable)

B) Location of the transformer

Name of the area where the transformer is located......

Condition of the area: Arid/semi-arid/medium potential/high potential region

Population pattern in the area densely populated/sparsely populated

C) Ownership of the transformers......

104