The list of words read out is in a random order, so that the subjects in Condition B (no category headings) couldn’t recall them as grouped categories. Cheating wasn’t an extraneous variable as it was made sure that the subjects were all spaced evenly so that they couldn’t see each other’s answers. Personal variables, such as reading, were considered and eliminated as the subjects were taken from an English class. The experiment was carried out in a quiet room so that noise didn’t become an extraneous variable.
Both of the conditions were run at the same time so there was no difference between each condition except from the independent variable. Also the sheets of paper were placed face down in front of the subjects, they all started and finished recalling words at the same time. So the amount of time allowed recalling the list of words didn’t affect the results. Measurement and Analysis To measure what affect the independent variable was having (category headings) the number of correct words recalled by each condition and subject were counted, this being the dependent variable.
The measurement technique was reliable as a pre-test was carried out on individuals that weren’t involved in the experiment and appropriate adjustments were made to the procedure. When all results were obtained the results from the two different conditions were compared in order to find the difference between the two conditions. In order to improve the comparative analysis of the data the results were then illustrated in the form of a comparative scatter graph and a pie chart.
Once the analysis had been carried out the Mann-Whitney Statistical test was used to determine whether there was a significant difference between the two conditions. Ethics The subjects that took part in the experiment were all above the age of consent (16 years) so parental consent was not necessary. Before the experiment began standardized instructions were read out to explain to the subjects that if they felt they were being put under any stress or pressure they could leave the experiment at any time.
(See Appendix A for the precise wording of the standardized instructions. ) The subjects were also debriefed at the end of the experiment to eliminate any feelings of failure. (See Appendix D for the precise wording of the debriefing. ) Results For the experiment each subject had their own sheet to record the list of words onto. In Condition A the subjects had printed category headings on the sheet and Condition B had a blank sheet to record the words onto. (Refer to Tables 1 and 2, Appendix E for the tables showing the subjects raw scores. )
Table 3 is a summary table showing the mean, medium and mode of the number of words recalled correctly for each condition. Condition A Category Headings Condition B No Headings Mean 11. 9 10. 5 Medium 11 10 Mode 10 10 Table 3. From table 3 it can be seen that the scores from each condition are very similar. This is shown through the mode as both conditions recalled 10 words correctly. With both the mean and the medium Condition A recalled more words than Condition B. The comparative scatter graph doesn’t show a positive or a negative correlation.
All that can be seen is that each condition has recalled a similar amount of words correctly, with Condition A having slightly higher points on the graph (therefore more words recalled correctly) than Condition B. A pie chart to show the mean value of each condition. This pie chart shows the data as descriptive statistics. It’s easier to compare the different conditions results through visual diagrams. You can see that Condition A’s mean value is slightly larger than Conditions B’s.
In order to find out how significant the difference between the two totals of Category Headings and No Headings, allowing a firmer conclusions to be made, a statistical test was carried out on the data. The Mann-Whitney Statistical Test was used due to the two conditions being unrelated. (See Appendix F for the calculations used. ) The value of U was found to be 68, which is more than the critical value for U at the significance level of 0. 05 for a one-tailed test. In conclusion there wasn’t a significant difference found between the mean scores for the two conditions.