There Are Two Main Types of Research Classifications

Research Studies

There are two main types of research classifications:

 Experimental – The researcher ______influences events and investigates the effects of the intervention. For example, clinical trials and laboratory experiments.

o These are often used when we are interested in studying the effect of a treatment on individuals or experimental units.

o Experiments are conducted when it is (ethically, physically, etc.) possible for the experimenter to determine which experimental units receive which treatment.

o Experiment Terminology

. Experimental Unit –

. Treatment –

. Response –

o Examples: Experimental Unit Treatment Response

o Only a well-designed and well-executed experiment can reliably establish ______.

 Observational – The researcher collects information on ______or ______of interest, but does not influence the results.

o The researcher compares the units that happen to have received a particular treatment of possess a particular characteristic of interest.

1 o Examples: Unit Treatment Response

. You cannot set up a control (non-smoking) group and a treatment (smoking) group.

o An observational study is useful for identifying ______cause of effects, but it cannot reliably establish causation.

Questions: Determine whether each of the following scenarios is an example of an experiment or an observational study. Explain.

1. In a test of roughly 200 men and 200 women, those with moderately high blood pressure (averaging 164/89 mmHg) did worse on tests of memory reaction time than those with normal blood pressure.

2. An examination of the medical records of more than 360,000 Swedish men showed that those who were overweight or who had high blood pressure had a higher risk of kidney cancer.

3. Over a 4-month period, among 30 people with bipolar disorder, patients who were given a high dose (10 mg/day) of omega-3 fats from fish oil improved more than those given a placebo.

4. Among a group of disabled women aged 65 and older who were tracked for several years, those

who had a vitamin B12 deficiency were twice as likely to suffer severe depression.

5. The leg muscles of men aged 60 to70 were 50% to 80% stronger after they participated in a 16- week, high intensity resistance-training program twice a week.

2 Types of Experimental Designs

 Completely randomized design – The treatments are allocated completely by random chance to the experimental units.

Example: Which of two varieties of tomatoes (A and B) yield a greater quantity of market quality fruit?

Some factors that may affect yield:  Different soil fertility  Exposure to wind/sun  Soil pH levels  Soil water content

Suppose we have the following field with 8 plots to plant tomatoes in. We need to randomly allocate the varieties to the plots. Since there are 2 varieties, how many plots will each variety get? ______

Question: Now, suppose the field sloped upward from left to right. Would this change how you assigned varieties to plots? Discuss.

3 Note: Randomization is an attempt to make the treatment groups as similar as possible, however we can only expect to achieve this when there is a large number of experimental units to choose from.

 Blocking – Group (block) experimental units by some known factor and then randomize within each block in an attempt to balance out unknown factors.

o Use blocking when:

. There are known factors (e.g., slope of the hill in the previous example)

. Randomization for unknown factors to try to “balance things out.”

Example: Multi-Center Clinical Trial Suppose a Mayo clinical trial comparing two chemotherapy regimens in treatment of patients with colon cancer will be conducted using cancer patients in Scottsdale, AZ and Rochester, MN.

Scottsdale Rochester

Question: How should the treatments be allocated to the 12 patients?

Example: Comparing Three Pain Relievers for Headache Sufferers

. What factors might the researchers want to control for in this study?

. How could blocking be used to increase precision of a designed experiment to compare the pain relievers?

. What are some other design issues?

4 Example: Comparing 17 Different Leg Wraps Used on Race Horses

. 17 “boots” were tested and each boot was tested 5 times. . Due to time constraints, all boots were not tested the same day. . 8 boots were tested on the 1st day, 5 on the 2nd day, and 4 on the 3rd day. . The leg was placed in the freezer and then thawed before the 2nd and 3rd days of testing.

Question: What problems do you foresee with this experimental design?

The following plot provides the force readings when no wrap or boot was used:

Question: What are the implications of these results?

Below, are the final boot comparisons:

5 Question: What should have been done? Example: Cholesterol Drug Study Suppose we wish to determine whether a drug will help lower the cholesterol level of patients who take it.

Question: How should the study be designed?

Example: Salk Polio Vaccine Field Trial In 1954 the Public Health Service organized an experiment to test the effectiveness of Salk’s polio vaccine.

o Polio – an epidemic disease with cases varying considerably from year to year. A drop in polio after vaccination could mean either: an effective vaccine or no epidemic that year.

o Subjects: 2 million children in 1st, 2nd, and 3rd grades.

o 500,000 were vaccinated (treatment group) – 2nd graders

o 1 million were deliberately not vaccinated (control group) – 1st and 3rd graders

o 500,000 were not vaccinated (parental permission denied) – Control group

Question: Discuss the flaws of this study.

6 Here is another version of the study.

o The control group was only chosen from students who had parental consent.

o Children were randomly assigned to either the treatment or control groups.

o A placebo was given (an injection of salted water was given to the control group).

o Diagnosticians were not told which group the subject came from.

The previous example utilized some key experimental design methods when conducting research on human subjects.

o Control group – a group used for comparison which does not receive a treatment or receives the previous treatment

o Blinding – subjects don’t know what treatment they are getting

o Double blind – subjects and administrators/diagnosticians don’t know which treatment was given/received

o Placebo – a dummy treatment (sugar pill, salted water injection, etc.)

Placebo Effect – when placebos are used in a study, there is often concern about the placebo effect. This is when people believe they are getting better or have been treated when they really haven’t. Approximately 35% of people respond positively to a placebo treatment.

Types of Observational Studies There are two major types of observational studies.

o Prospective -

Example:

o Retrospective –

7 Example:

Comments:

 Observational studies should use some form of random sampling to obtain representative samples

 Observational studies cannot reliably establish causation.

Example: A prospective study was carried out over 11 years on a group of smokers and non-smokers showed that there were 7 lung cancer deaths per 100,000 in the non-smoker group and 166 lung cancer deaths per 100,000 in the smoker group.

Question: What other factors could have influenced these results besides smoking?

The researchers could try to control for some of these ______factors.

Example: “Home births give babies a good chance” (NZ Herald, 1990) An Australian report was stated to have said that babies are twice as likely to die during or soon after a hospital delivery than those from a home birth. The report was based on a simple random sample of home births and hospital births.

Question: Does this mean hospitals are dangerous places to have babies in Australia? Explain.

Example: “Lead exposure linked to bad teeth in children” (USA Today) The study involved 24, 901 children ages 2 and older. It showed that the greater the child’s exposure to lead, the more decayed or missing teeth they had.

Question: Does this show lead exposure causes tooth decay in children? Explain.

8 The researcher in the article said “We controlled for income level, the proportion of diet due to carbohydrates, calcium in the diet, and the number of days since last dental visit.”

Example: Suppose we are interested in conducting a study to determine whether age at 1st pregnancy is a risk factor for cervical cancer. How might we proceed?

Example: Suppose we are interested in conducting a study to determine if sentencing of convicted felons is racially biased. How might we proceed?

Example: Suppose we are interested in conducting a study to determine if a fruit juice supplement reduces muscle pain following exercise. How might we proceed?