The mean shows that on average the participants in the chocolate condition rated their happiness a 4, which is higher than the average happiness score of those in the no chocolate condition. This indicates that those in the chocolate condition felt slightly happier than those in the no chocolate condition. This is also supported by the Mode, as the mode for group 1 was higher than for group 2. The median is the most suitable measure of central tendency for this study as the data collected was ordinal. This also showed that generally group 1 rated their happiness higher than group 2. A higher median indicates that group 1 had higher happiness scores overall.
In order to find out whether the results were due to chance factors a Mann-Whitney U test was carried out. The observed value of U was 52 and the critical value was 64 at a 0.05 level of significance. As the observed value of U was smaller than the critical value the experimental hypothesis was accepted and the null hypothesis was rejected (Appendix 6, Page 10). The graphs show a range of answers from the two groups. In group one the happiness scores range from 3 to 5 and in group two the happiness scores range from 2 to 4. From the graphs it can be seen that majority of subjects in both of the groups rated their happiness 3 or 4. Showing little variation between the two groups.
The results support the experimental hypothesis stating subjects in the chocolate condition will significantly score their happiness higher than those in the no chocolate condition. There was a lot of control over the study as an experiment was used and the independent variable was completely controlled. This research method also ensured that all subjects were treated equally as possible. They all heard the same debriefing but the standardised instruction differed for the two groups, as it involved one group eating chocolate and the other not.
A strength of using the independent measures design is that each subject took part in only one of the two conditions avoiding order effects. But participant variables may have been a problem for example the participants in the chocolate condition may have been more comfortable with sharing their happiness score than those in the other group. The sample included equal number of males and females; therefore the study was not biased towards gender. But the results cannot be generalised to people outside the age range 16-19 years and to those who are not students. Also, the results don’t represent other towns and can only be generalised to British students. Therefore many generalisations cannot be made.
Demand characteristics may have occurred as the subjects were informed in the standardised instructions that they will be answering questions. There is also the possibility that subjects may have rated their happiness on the basis of what they thought was expected rather than how happy they really were at the time. But demand characteristics were also minimised by asking other questions rather than just the critical question.
To improve the study, subjects from different towns, nationalities and age groups could be used to see if similar results are found and this would enable the findings to be generalised to a much wider population. Subjects: An opportunity sample of 30 black and white college students aged between 16-19 years. 15 males and 15 females. Subjects were selected during college time and randomly assigned into one of the two groups.
Procedure: Informed consent was gained from all participants and they were checked for any allergies or illness that may be affected by chocolate (Appendix 1, Page7). The study took place in a quit room, where only the participants were present. The standardised instructions were read to both groups (Appendix 2, Page 7). Group 1 (chocolate condition) did the experiment first. All 15 participants in that group were provided with 4 squares of chocolate. After they ate the chocolate, they filled in the short questionnaire. Then they left the room and immediately group 2 came in and filled in the same questionnaire but they were not given any chocolate. Both of the groups were thanked and debriefed (Appendix 3, Page 8).
All participants were treated equally. The experiment took place in the same room ensuring identical surroundings. All subjects were aware that they had the rights to withdraw at anytime (Appendix 1 & 2, Page 7). The only difference was that one group were given chocolate before they filled in the questionnaire but the other group was not. The dependant variable was the answer to: Rate your happiness on the scale of 1-5. This was the critical question; the other questions were there just to reduce demand characteristics. In order to find out whether the results were just due to chance an inferential statistical test was carried out. The experiment used an independent groups design; therefore the Mann-Whitney U test was used at a 0.05 level of significance. A significant difference between the happiness of the two groups was looked for.
It was found that the subjects in the chocolate condition rated their happiness higher than those in the no chocolate condition. This indicates that chocolate does have an affect on a person’s happiness. The table below shows the mean, median & mode. They have been used to find out the central tendency for the happiness scores for both of the groups. (See Appendix 5, Page 9 for raw data).