An Examination of Affective Forecasting by Basketball Players

Home , Desire, Disappointment

CALIFORNIA STATE UNIVERSITY, NORTHRIDGE

MISPREDICTIONS: AN EXAMINATION OF AFFECTIVE FORECASTING BY

BASKETBALL PLAYERS

A thesis submitted in partial fulfillment of the requirements

For the degree of Master of Arts in

Psychology, General-Experimental

Raffi Sarafian

August 2015

The thesis of Raffi Sarafian is approved:

______Date:______

Dr. Jacob Jensen

______Date:______

Dr. Debra Berry Malmberg

______Date:______

Dr. Mark P. Otten, Chair

California State University, Northridge

ii Acknowledgement

I would first like to thank my graduate advisor, Dr. Mark P. Otten, for his guidance, support and mentorship over the past three and a half years. I would also like to thank Drs. Ashley Samson, Jacob Jensen and Debra Malmberg for serving on my thesis committee and providing feedback, which was crucial to the success of this project.

Additionally, this idea came to me through my competing in bicycling races and constantly feeling as though I wasn’t performing to my satisfaction as I always had hoped. I would like to thank my parents for their support through every step of this project, from the conception of the hypothesis through the final draft.

iii TABLE OF CONTENTS

Signature Page ii

Acknowledgement iii

Abstract v

Introduction 1

Current Study 8

Method: Study 9

Participants 9

Measures 9

Procedure 10

Results: Study 12

Discussion 14

Limitations and future Direction 16

Conclusion 17

References 19

Appendix 22

iv ABSTRACT

MISPREDICTIONS: AN EXAMINATION OF

AFFECTIVE FORECASTING BY BASKETBALL PLAYERS

Raffi Sarafian

Master of Arts in

Psychology, General-Experimental

Previous research on affective forecasting has investigated people’s perceptions of feelings and the way they predict their future emotional states. This is seen in every day situations by anyone and within athletes on teams. Research has revealed that for a team or group to succeed they must use team cohesion tactics. Team cohesion had not previously been studied as a covariate of affective forecasting to decrease mispredictions of feelings towards the outcome of future events. The present study hypothesized that basketball athletes would mispredict their feelings towards the outcome of a game regardless of that outcome. The present study also hypothesized that according to the forces of team cohesion, athletes playing on a team would predict future emotional states more accurately than those playing individually. Participants (n=115) with previous basketball playing experience were randomly assigned to either a one vs. one or two vs. two basketball playing condition. The participants completed a pre-game questionnaire, played a ten-minute basketball game, and completed a post-game questionnaire. Results revealed that regardless of the outcome of the game, athletes mispredicted their future emotional state toward the game's outcome. Results also showed that participants who played on teams predicted their future emotional states more accurately than those

v playing individually. These findings suggest that team cohesion can lead to a better success in predicting one’s own future emotional state.

vi INTRODUCTION

Anyone who has visited the Las Vegas sports book, poker room, or craps table knows that predictions, whether correct or incorrect, are something that can bring about great emotion. One might experience happiness and excitement, or depending on the outcome of the prediction, anger and discomfort. Research on prediction shows that people tend to mispredict or overestimate their emotional state towards the outcome of a future event, which has been defined as affective forecasting (Wilson & Gilbert, 2005). Working in teams or a group could potentially decrease the likelihood of making this misprediction of emotional states. How closely a team or group works together has been defined as group cohesion (Festinger, 1950). To date, there is little research on group cohesion and its relationship with affective forecasting -- that is, the effect of cohesion on the accuracy of one’s prediction of future feelings toward the outcome of an event.

Affective Forecasting and Impact Bias

To further understand the mechanisms behind affective forecasting, one must understand the biases associated with it. Most commonly observed has been the impact bias, in which those who are forecasting mispredict the intensity of their predicted response to a future event. Kermer, Driver, Wilson, and Gilbert (2006) investigated affective forecasting and impact bias. In their study, 54 participants played a gambling computer game; participants had to guess which suit card was the top-ranked according to the computer. Participants were awarded 50 cents every time they guessed a correct top-ranked suit and 25 cents when guessing the second-ranked; if they guessed they lost 25 cents for the third-ranked and lost 50 cents if they guessed the fourth-ranked suit card. After the 25th

1 trial, the amount they won or lost so far was doubled. Both before and after the game, participants reported their level of happiness or unhappiness. The findings were that the participants who lost money experienced a greater impact on emotion than when they gained money. The researchers concluded that even if people have experienced losses throughout their lives, they still tend to mispredict the intensity of a loss compared to a gain in the same type of situation (Kermer et al., 2006).

This misprediction has also been documented by other researchers. Morewedge and

Buechel (2013) conducted a study in which participants were given pictures of two professors and told that they would compete against them in a simple motor skills task (i.e., pressing a two-letter sequence on the computer). Participants were randomly assigned to one of two groups. The participants in the first group were given a choice of which professor they wanted to compete against; those in the second group were told that they were competing against both professors. Prior to attempting the task, participants in both groups were asked to forecast the level of happiness they thought they would feel if they were to beat their chosen professor (for group 1) or both professors (for group 2). In both conditions, participants were told that they were the winners but were not told the exact score the professor(s) received, but the participants were always the winners. Results showed that those who were given the option to choose which professor they wanted to beat before actually attempting the task were more likely to commit the impact bias and mispredict their intensity of happiness.

Morewedge, Gilbert, and Wilson (2005) investigated whether people could better forecast and predict their emotions towards the outcome of an atypical and highly

2 memorable event. The researchers’ hypotheses were: 1) when people forecast their feelings towards the outcome of a future event, they will rely on atypical instances (they are not accustomed to) and 2) when unaware that these instances are atypical, people will rely on affective forecasts. In study one, participants were male and female passengers getting off of a train. Participants were randomly assigned to one of three conditions which were to describe a situation in which they missed a train (free recall condition), to describe their worst instance of missing the train (biased recall condition), or to describe three instances of missing their train (varied recall condition). Each participant then reported their level of happiness. Results showed that participants in the free recall condition made more negative forecasts than those in the biased and varied recall conditions. Overall, when a person is asked to freely describe a situation, there is a great deal of misprediction than when it relates to a much more specific event.

In Morewedge and colleagues’ (2005) second study, the concept was kept the same as the first study but conducted during an undergraduate football game. The sample was 54 football fans randomly selected prior to the start of the football game. The participants were randomly assigned to one of three conditions and were asked to describe either a football game they had seen when the team won (free recall), the best football game they had seen the team win (biased recall), or describe three football games where the team won

(varied recall). The results revealed that there were more positive forecasts in the free recall condition than in the biased or varied recall conditions which were more negative.

In another study exploring misprediction, Gilbert and Ebert (2002) aimed to see if people prefer changeable outcomes or unchangeable outcomes and if they can predict the

3 consequences of an outcome if changed. The researchers hypothesized that those who have been given an option to change their choice of a picture and given a few days to think about it will have greater impact bias in forecasting. In the study, participants from a photography class were randomly assigned to one of two groups (changeable or unchangeable groups) and asked to find two pictures that they really liked but could only keep one. The participants in the unchangeable condition could not go back to switch the picture after choosing one because the other picture would be thrown away. In the other condition, the participants were allowed to change their current selection of artwork with the one they did not select. The results revealed that when faced with a few days to change the picture, a participant experiences a longer period of wanting to change the picture and also mispredict their emotional levels.

Affective Forecasting and Sports

Researchers have also investigated affective forecasting in athletes. Research done by Dijk (2009) intended to determine whether track athletes experience the impact bias.

The study investigated two hypotheses:1) according to the impact bias, track athletes would overestimate the intensity of their emotions following a race regardless of their success at the meet, and 2) according to the impact bias, track athletes would also predict that a loss impacts their emotions more than a win. Participants were professional and semi-professional track athletes competing in a track meet. A Likert scale from 1-9 was used assessing the current state of happiness experienced before and after the race, developed by Wilson & Gilbert (2005). The researcher approached the athletes before a track race and asked them to complete a questionnaire measuring what they predicted their

4 emotions would be following the race, and another survey once again after the race measuring their current emotional states (Dijk, 2009). The results showed that athletes mispredicted their emotional levels and intensities and that impact bias was greater when losing than when winning a race. On average, athletes who won overestimated their level of intensity of positive emotions, and athletes that lost overestimated their level of intensity of negative emotions. Findings suggest that the knowledge of an athlete’s mispredictions could assist in thinking positive thoughts in order to be successful, but can also set up another athlete for failure if he/she makes negative forecasts about the outcome of an event causing unnecessary dread and doubt.

In a recent study by Loehr and Baldwin (2014), affective forecasting was investigated among individuals who exercise. The researchers predicted that those who actively participate in daily physical activity would predict their future emotional states towards the outcome of a workout better than those who do not regularly exercise.

Participants met the researchers at the university gym and were asked to write out a workout plan for thirty minutes; they were then given a questionnaire asking them to predict what they thought their emotional state would be after their workout. After participants completed the workout, they completed the questionnaire again regarding their current emotional state. The results revealed that all participants enjoyed the workout more than they had expected to, on average. Results also showed that those participants who did not report regular participation in physical activity had larger gaps in their affective forecasting predictions (pre- to post-workout) than those who did regularly participate in

5 physical activity. Overall, these studies show that affective forecasting can be a vital aspect for athletes playing individually; this is also the case with athletes playing on a team.

Team Cohesion in Sports

Success in team sports is often dependent upon group cohesion. Festinger (1950) suggested that the two forces that drive group cohesion are the benefits that come from working with a group to achieve a common goal and a desire for interpersonal interactions within the group. Further research by Carron, Brawley, & Widmeyer (1998) affirmed the two versions of group cohesion for sports teams. One version is task cohesion, which can be described as the degree to which group members work with one another to strive towards a common goal. The second version of group cohesion is social cohesion – the degree to which the group members enjoy each other’s company when working together.

Carless and DePaola (2000) examined whether putting workers in an office setting into teams increased work productivity. The researchers required employees from a retail store to work in groups; after working in groups, participants completed questionnaires on team effectiveness, team cohesion, job satisfaction, and work-group performance. Results showed that, of all of these factors, task cohesion was the most important in the effectiveness of the productivity of the group.

A similar finding can also be seen within sports teams. Ramesh (2011) examined the level of performance and amount of cohesion within a team and hypothesized that through positive group cohesion, such as working together and properly communicating during a game, volleyball players’ team performance would increase. Four male volleyball teams of 12 players each were administered questionnaires assessing the interaction,

6 attraction, integration, and similarity within their team. The teams participated in a tournament in which there were two winning teams and two losing teams. The researchers found a significant correlational relationship between the cohesion of the volleyball players

(e.g., better interactions with one another) and how well they performed during a game.

In a summary of literature on team cohesion and sport, Carron, Colman, Wheeler and Stevens (2002) conducted a meta-analysis consisting of 46 studies examining team cohesion and performance. They found that for the majority of the studies, working closely as a team improved performance on the same types of tasks or skills test as compared to working individually. Results also showed that those teams that had positive relationships and worked strongly closely together performed better than those who did not.

Affective Forecasting and Team Cohesion

Mallett, Wilson, & Gilbert (2008) further examined the intensity levels of people’s forecasted emotional states within groups. Participants were randomly assigned to two different groups in which they were asked complete a task with another participant from the other group; this occured over five days. Before each interaction, participants were asked to rate how they expected the interaction would go and to journal the interaction afterwards. Results revealed that participants’ forecasts before the interactions were more negative than their diary reports after the interactions. Thus, participants mispredicted the intensity of their negative emotional states towards those in their outgroup.

As research shows, group cohesion is highly important for success in team sports; meanwhile, athletes may fall victim to mispredictions with affective forecasting. It is unclear whether team and group cohesion impacts affective forecasting – that is, if athletes

7 in team sports can more accurately predict their feelings towards outcomes of future events.

Research has not yet explored the relationship between affective forecasting and team cohesion. Research on team cohesion suggests that if athletes work with one another on the same level, then they might begin to anticipate and predict one another’s actions and emotional states. Given this feedback from teammates, an athlete might then be able to predict the outcome of his/her own game more accurately. For example, a basketball player might notice confidence and a lack of anxiety in her teammate, and given high cohesion on the team, might then more accurately predict her own successful performance.

Hypotheses

The purpose of this study was to examine the effects of affective forecasting and its interaction with team cohesion for basketball players. The research hypotheses are: (1) basketball players would mispredict their level of happiness towards the outcome of a future basketball game they participate in, regardless of the outcome, and (2) building upon the ideas behind team cohesion and the meta-analysis done by Carron et al. (2002), the players would predict their emotional state more accurately when playing on a team as opposed to when playing individually.

8 METHOD

Study

Participants

One hundred and fifteen (n = 115) male California State University, Northridge undergraduate students between the ages 18-25 were participants for this study.

Convenience sampling was used to gather participants from the psychology department’s research pool. Participants’ ethnic backgrounds were: 50.4% Hispanic, 18.3%

Asian/Pacific Islander, 12.2% African American, 10.4% Caucasian, 8.6% multi- ethnic/other. To sign up for the research study, participants were required to be male

(females were excluded for this study as it was an initial study into affective forecasting and team cohesion)to have experience playing on a high school and/or recreation/intramural basketball team. Before signing up for the study online, participants were required to complete the Physical Activity Readiness Questionnaire (PAR-Q) with confirmation of “no” on all items. This qualified a student as being in a physically healthy condition and able to run and jump to participate in the study. The average number of years played on a competitive team by all participants was 4.92 years (SD =9.95; range = 1-10)).

The participants' average age was 19.91 years old (SD = 2.05; range: 18-25). No monetary incentive was given for participating, but the participants received research participation credits as compensation. The participants are referred to “basketball players” for the rest of the study.

Measures

9 The Affective Forecasting Scale (Gilbert & Ebert, 2002) measured athletes’ forecasted negative and positive emotions on 7-point Likert-type scale ranging from 1 (not happy) to 7 (very happy). Negative emotions were assessed by rating the disappointment, negative feelings, and frustration they would feel if they lost the game. Positive emotions were assessed by rating the happiness, positive feelings, and satisfaction they would feel if they win the game.

The Group Environment Questionnaire (GEQ; Carron, Widmeyer & Brawley,

1985) was used in the team condition to assess cohesion. This survey consists of consists of

18 items assessing individual attractions to the group within the task, group integration and social aspects of the task. Responses are based on a self-report 9-point Likert-type scale ranging from 1 (strongly disagree) to 9 (strongly agree).

Procedure

Participants were randomly assigned to either the individual or team condition. The individual condition had two participants at a time. The participants entered the basketball gymnasium and were greeted by the researcher. A researcher handed them the pre-game questionnaire, administered on an iPad, which consisted of the questionnaires detailed in the materials section. After the participants completed the questionnaires, they were instructed to play a basketball game together for ten minutes. One researcher kept track of the score, and the other acted as referee of the game (i.e., watching for fouls). After ten minutes, the researcher acting as referee ended the game and the two participants were given a minute to catch their breath and rest. Then they were given the post-game

10 questionnaire packet on the iPad, a repeat of the pre-game questionnaire packet. After completing the packet, they were debriefed and thanked for participating in the study.

The procedure for the team condition was the same,but included four participants at each time, who were randomly assigned to teams and played a two-versus-two game. The participants who were on the same team were allowed to talk to one another but were not permitted to speak with the participants who were on the other team. After the participants completed the post-game questionnaires, they were debriefed and thanked for participating in the study.

11 RESULTS

SPSS version 22 was used to run the analysis on the final sample which resulted in n = 115. All responses fell within the scale ranges, and thus no outliers were identified or excluded.

Analysis: Hypothesis 1

A series of dependent-samples t tests were run for hypothesis 1. Two composites were created; one was derived from the pre-game affective forecasting scale, and the other was a similar post-game affective forecasting composite. For the pre-game composite,

Cronbach’s α was .56. To improve this reliability, the fifth item (“How happy do you think you will immediately feel if you lose?”) was excluded, boosting the αvalue to .68.

Reliability of the four-item post-game version of the scale was similar, Cronbach’s α= .69.

Based on a comparison of these composites, the analysis revealed that players did indeed mispredict their future emotional state across both individual and team conditions regardless of the game’s outcome (pre-game: M = 5.39, SD = 0.91; post-game M = 4.50,

SD = 1.13), t (114) = 7.51, p < .001, Cohen’s d = .70.

Since the above composite reliabilities were not particularly strong, Item 1 on the affective forecasting pre-game questionnaire was selected and compared with its corresponding item on the affective forecasting post-game questionnaire, as follows. A dependent-samples t test found a significant mean difference when comparing the pre- game item “How happy do you think you will feel immediately afterwards?” (M = 5.20,

SD = 1.20) with the post game item “How do you currently feel?” (M = 4.72, SD =1.54), t

(114) = 2.84, p = .01, Cohen’s d = .27. See table 1 in Appendix.

12 Analysis: Hypothesis 2

Two dependent-samples t tests were run to test for hypothesis 2. In the individual condition, a significant difference was found when comparing the pre-game item “How happy do you think you will feel immediately afterwards?” (M = 5.17, SD = 1.34) with the post-game item “How do you currently feel?” (M = 4.51, SD =1.61), t (46) = 2.23, p = .03,

Cohen’s d = .33. However, another dependent-samples t test revealed no difference for players in the team condition when comparing the same two items, pre-game (M = 5.22,

SD = 1.10) versus post-game (M = 4.87, SD = 1.50), t (67) = 1.77, p = .08. See table 2 in

Appendix.

Analysis: Group Environment Questionnaire

The results gathered from the Group Environment Questionnaire can be seen in table 3 of the Appendix section showing the means and standard deviations of the items observed. This was used as a means to see if the players were actually co-operating and working well with another in which results from the questionnaire revealed so.

13 DISCUSSION

The purpose of the present study was to examine affective forecasting and whether basketball players mispredict their future emotional state towards the outcome of a basketball game. The first hypothesis was that basketball players will mispredict their future emotional state towards the outcome of a game regardless of the actual outcome.

The results suggested that basketball players did mispredicted their future emotional states no matter the outcome of the game or condition. The findings of the experiment fully supported the hypothesis and were consistent with the base findings of affective forecasting from Kermer et al. (2006). The purpose of their work was to find a difference in individuals’ predictions of their future emotional states towards the outcome of tasks such as gambling. The current study examined basketball players with the same purpose set forth and found similar results. The findings of this study support that the principles of affective forecasting and misprediction of future emotional states can also be applied for basketball players going into a game. The affective forecasting scale used was the same as the one developed by Kermer et al. (2006) for a gambling task. Similar to the studies investigating affective forecasting with runners mispredicting their future emotional state towards the outcome of a track meet (Dijk, 2009) and exercisers who predicted future emotional states towards exercising more accurately than those who do not regularly exercise (Loehr & Baldwin, 2014), results were similar in that basketball players also mispredict their future emotional states regardless of the outcome of the competition.

The most important finding from this study was when examining the second hypothesis. The second hypothesis was that basketball players playing on teams will

14 predict their future emotional states towards the outcome of the game more accurately than those playing individually. Studies focusing on team and group cohesion stemming from

Festinger’s (1950) original definition and conception of team cohesion and Carron and colleagues' (2002) meta-analysis of team/group cohesion were taken into account combined with affective forecasting to build a new and original hypothesis. The results supported the team cohesion findings from Carron et al. (2002) along with Kermer et al.

(2006)’s findings to establish the first relationship between affective forecasting and team cohesion. Scores from the pre- and post-game questionnaires of the individual and team conditions were compared. There was a significant pre- to-post-game difference of mean scores in the individual condition, meaning participants mispredicted their future emotional states. However, in the team condition, there was no significant difference between pre- and post-game questionnaire scores, which in this case is exactly what was predicted. This implies that basketball players in the team condition predicted their future emotional states more accurately than the players in the individual condition.

The current study was unique and first of its kind to examine the relationship between affective forecasting and team cohesion. Looking at previous research by Mallet

(2008), in which groups mispredicted future emotional states towards outgroup interactions, as well as team cohesion studies by Ramesh (2011) in which volleyball players had more success working well with one another, and Carless and DePaola

(2000)’s office workers’ better success when working together, the present hypothesis provides a more unique take on the idea of team cohesion. Instead of concluding that team

15 cohesion leads to greater success, in the present study, team cohesion leads to predicting one’s future emotional state more accurately.

The data from the Group Environment Questionnaire suggests that people felt team cohesion during the basketball game in this study. For example, when examining item 7, which asked, “Our team is united in trying to reach its goals for performance” had a mean of M=6.68, meaning that the players actually felt as if their teammates were also trying hard to reach their goal and suggesting team cohesion. Also, item #8 was, “We all take responsibility for any loss or poor performance by our team”, which had a mean of

M=6.99, meaning the players agreed that their teammates also take some of the responsibility for a loss, once again showing team cohesion.

Limitations and Future Directions

A limitation within this study was the sample and the use of the basketball players who were less experienced than those on an intercollegiate or professional level. A potential follow-up study could examine the same hypotheses among these college or pro basketball players, which might show a stronger effect for team cohesion and affective forecasting due to the players already having a solid team foundation.

Another possible limitation was with the affective forecasting scale. There were two items relating to the game’s outcome on the pre- and post-game questionnaires that differed between their pre- and post-versions. The pre-game item, “How happy do you think you will immediately feel if you win?” compared with post-game item, “If you won the game, do you feel as happy as you thought you would?” and pre-game item, “How happy do you think you will immediately feel if you lose?” compared with post-game item

16 “If you didn't win the game, do you feel as happy as you thought you would?” were left out of the analysis. Most participants answered both “If you won the game” and “If you lost the game” questions on the post-game questionnaire. This created a discrepancy when analyzing the data because it was not clear which participants had won or lost the game.

This limitation is not a setback because in the hypothesis it is stated that we are looking for a result regardless of the outcome of the game.

Another limitation within this study may be that players in the team condition did not know one another. A possible friendly greeting and conversation exchange might assist with this limitation for future studies; however, the data showing team cohesion show results supporting the hypothesis.

Future studies could examine these effects in female athletes as well. Research could also be expanded to examining more than just basketball, but also other team sports such football, baseball, soccer and hockey. Alternatively, one could test if players on teams of more than two would make a more accurate prediction of emotional states, given perhaps even years of experience interacting with one another within teams. Through the studies done by Ramesh (2011) and Carless and DePaola (2000), the conclusion can be made that a team working efficiently together with rapport with one another can lead to greater success which can also possibly lead to a stronger prediction.

Conclusion

Ultimately, the goal of this study was to analyze affective forecasting among basketball players and see whether the players mispredict their future emotional states towards the outcome of a game. Results supported the hypothesis and found that basketball

17 players do mispredict their future emotional states regardless of the outcome of a game.

Our second hypothesis asked whether basketball players on teams will more accurately predict their future emotional state than those playing individually; results also supported this hypothesis. Now that we understand and are more aware of how we mispredict our future emotional states, perhaps during the next visit to Las Vegas to gamble, one should do as the expression says, “leave all your emotions at the door”.

18 References

Carless, S. A., & De Paola, C. (2000). The measurement of cohesion in work teams.

Small Group Research, 31(1), 71-88. doi:10.1177/104649640003100104

Carron, A. V., Brawley, L. R., & Widmeyer, W. N. (1998). The measurement of

cohesiveness in sport groups. In J. L. Duda (Ed.), Advances in sport and exercise

psychology measurement (pp. 213-226). Morgantown, WV: Fitness Information

Technology

Carron, A. V., Colman, M. M., Wheeler, J., & Stevens, D. (2002). Cohesion and

performance in sport: A meta analysis. Journal of Sport & Exercise Psychology,

24(2), 168-188.

Carron, A.V., Widmeyer, W.N. and Brawley, L.R. (1985). The development of an

instrument to assess cohesion in sport teams: the Group Environment

Questionnaire. Journal of Sport Psychology, 7, 244-266.

Cox, R. H., Martens, M. P., & Russell, W. D. (2003). Measuring Anxiety in Athletics:

The Revised Competitive State Anxiety Inventory-2. Journal Of Sport & Exercise

Psychology, 25(4), 519-533.

Dijk, W. (2009). How Do You Feel? Affective Forecasting and the Impact Bias in Track

Athletics. Journal of Social Psychology, 149(3), 243-248. Festinger, L., Schachter,

S., & Back, K. (1950). Social pressures in informal groups; a study of human

factors in housing. Oxford England: Harper.

Gilbert, D. T., & Ebert, J. J. (2002). Decisions and revisions: The affective forecasting of

19 changeable outcomes. Journal of Personality And Social Psychology, 82(4), 503-

514. doi:10.1037/0022-3514.82.4.503

Gilbert, D. T., Pinel, E. C., Wilson, T. D., Blumberg, S. J., & Wheatley, T. P. (1998).

Immune neglect: A source of durability bias in affective forecasting. Journal Of

Personality And Social Psychology, 75(3), 617-638. doi:10.1037/0022-

3514.75.3.617

Kermer, D. A., Driver-Linn, E., Wilson, T. D., & Gilbert, D. T. (2006). Loss aversion is

an affective forecasting error. Psychological Science (Wiley-Blackwell), 17(8), 649-

653. doi:10.1111/j.1467-9280.2006.01760

Loehr, V. G., & Baldwin, A. S. (2014). Affective forecasting error in exercise:

Differences between physically active and inactive individuals. Sport, Exercise, and

Performance Psychology, 3(3), 177-183. doi:10.1037/spy0000006

Mallett, R. K., Wilson, T. D., & Gilbert, D. T. (2008). Expect the unexpected: Failure to

anticipate similarities leads to an intergroup forecasting error. Journal of

Personality and Social Psychology, 94(2), 265-277. doi:10.1037/0022-

3514.94.2.94.2.265

Morewedge, C. K., Gilbert, D. T., & Wilson, T. D. (2005). The Least Likely of Times.

Psychological Science, 16(8), 626-630.

Ramesh, N. (2011). Relationship between team cohesion and performance among

university level male volleyball players. Journal of Arts and Culture,2(2),40-42.

Wilson, T. D., & Gilbert, D. T. (2005). Affective forecasting: Knowing what to want.

20 Current Directions In Psychological Science, 14(3), 131-

134.doi:10.1111/j.09637214.2005.0035

21 Appendix

Table 1

Hypothesis 1 Results

Composite Item 1

Pre-game M = 5.39, SD = 0.91 M = 5.20, SD = 1.20

Post-game M = 4.50, SD = 1.13 M = 4.72, SD =1.54

t (114) = 7.51, p < .001 t (114) = 2.84, p = .01

22 Table 2

Hypothesis 2 Results

Individual Team

Pre-game M = 5.17, SD = 1.34 M = 5.22, SD = 1.10

Post-game M = 4.51, SD =1.61 M = 4.87, SD = 1.50

t (46) = 2.23, p = .03 t (67) = 1.77, p = .08

23 Table 3

Group Environment Questionnaire Results

Individual Attractions to the Individual Attractions to the Group- Social (ATGS) Group-Task

(ATGT) Item# Score Item# Score

1 M = 3.63, SD = 2 M=5.24, SD = 2.48 2.46 3 M=3.10, SD = 4 M=2.26, SD = 2.18 2.10 5 M=3.85, SD = 6 M=3.07, SD = 2.52 2.18 7 M=6.68, SD = 8 M=6.99, SD = 2.03 1.84 9 M=4.62, SD = 2.38

24 Questionnaire 1

Affective Forecasting Scale

1. I have been smiling and laughing a lot. 1 2 3 4 5 6 7 Strongly Strongly Disagree Agree

2. My future looks good.

1 2 3 4 5 6 7 Strongly Strongly Disagree Agree

3. How happy do you think you will feel immediately afterwards?

1 2 3 4 5 6 7 Not very happy Very happy

4. How happy do you think you will immediately feel if you win?

1 2 3 4 5 6 7 Not very happy Very happy

5. How happy do you think you will immediately feel if you lose?

1 2 3 4 5 6 7 Not very happy Very happy

25 Questionnaire 2

Group Environment Questionnaire (GEQ)

This questionnaire is designed to assess your perceptions of your team. There are no wrong or right answers, so please give your immediate reaction. Some of the questions may seem repetitive, but please answer ALL questions. Your personal responses will be kept in strictest confidence.

The following statements are designed to assess your feelings about YOUR PERSONAL INVOLVEMENT with this team. Please CIRCLE a number from 1 to 9 to indicate your level of agreement with each of these statements.

1. I do not enjoy being a part of the social activities of this team.

1 2 3 4 5 6 7 8 9