Examination for BSc Psychology (SPECIMEN PAPER)
4PAHPRM2 RESEARCH METHODS 2 PRACTICAL EXAMINATION
Section A
The questions in Section A are about the article by Shalvi et al. (2012).
You are provided with information about a replication of Experiment 1 from this article. This information includes output from analyses done using SPSS.
Question A1 [11 marks available]
Shalvi et al. (2012) report this analysis on page 1266 of their article:
Time. As we expected, participants in the high-time-pressure condition took less time to report the outcome of the die role (M = 10.8 s, SD = 4.4 s) than those in the low-time-pressure condition (M = 23.2 s, SD = 11.2 s), t(70) = −6.01, p < .001, which indicates that the time-pressure manipulation was successful.
The equivalent analysis was done in SPSS for the replication data and is shown below in the boxes below:
(a) Use the SPSS output above to complete the paragraph below. Report data to the same degree of precision as used by Shalvi et al. (2012). For alternative statements shown in bold red font, delete to leave the most appropriate statement. Red lines indicate where information is to be inserted.
As expected / Counter to expectations participants in the high-time-pressure condition took less / more time to report the outcome of the die role (M = ___ s, SD = ___ s) than those in the low-time-pressure condition (M = ___ s, SD = ___ s), t(___) = ___ , p ___ , which indicates that the time-pressure manipulation was / was not successful.
(b) Is the result statistically significant? (Select one answer): Yes / No
(c) Which kind of t-test was used? (Select one answer):
|
One-sample
|
Related samples
|
Independent samples
|
(d) The outcome of this test has important implications for how other analyses in this experiment are interpreted. In this regard, briefly explain what the main implication of a non-significant outcome to this test would be. (Two or three sentences are sufficient)
[A1: 11 marks available]
Question A2 [12 marks available]
Shalvi et al. (2012) report this analysis of the low-time-pressure condition on page 1266 of their article:
The distribution of reported outcomes differed from the uniform. distribution expected from a fair die roll in [both] the low-time-pressure condition, χ2(5, N = 38) = 13.47, p = .02, …
The equivalent analysis was done in SPSS for the replication data and the output for the statistical test is shown in the box below:
(a) Table A2 shows the observed frequencies for this analysis. Complete the cells for the expected frequencies in Table A2.
Table A2. Expected frequencies if the data follow a uniform. distribution, and observed frequencies for the number reported as showing on the die.
|
|
Number (reported) showing on the die
|
|
1
|
2
|
3
|
4
|
5
|
6
|
|
Expected frequency
|
|
|
|
|
|
|
|
Observed frequency
|
3
|
4
|
5
|
7
|
6
|
8
|
(b) Use the SPSS output above to complete the paragraph below. Report data to the same degree of precision as used by Shalvi et al. (2012). For alternative statements shown in bold red font, delete to leave the most appropriate statement. Red lines indicate where information is to be inserted.
The distribution of reported outcomes differed significantly / did not differ significantly from the uniform. distribution expected from a fair die roll in the low-time-pressure condition, χ2( ___, N = ___) = ___, p ___.
(c) Which kind of chi-square test was used? (Select one answer):
Chi-square goodness-of-fit test / Chi-square test for contingency tables
(d) What do you conclude from this test? (One sentence is sufficient)
(e) With reference to the aim (purpose) of the research by Shalvi et al., briefly explain why testing the hypothesis that the data follow a uniform. distribution is the appropriate null hypothesis to test in this analysis. (Two or three sentences are sufficient)
[A2: 12 marks available]
Question A3 [5 marks available]
Shalvi et al. (2012) report this analysis that includes a Mann-Whitney U test on page 1266 of their article:
Supporting our hypothesis, participants in aggregate lied significantly more, by reporting higher die-roll outcomes, in the high-time-pressure condition (M = 4.56, SD = 1.48) than in the low-time-pressure condition (M = 3.87, SD = 1.61), Mann-Whitney Z = 2.03, p < .05.
The equivalent analysis was done in SPSS for the replication data and the output for the statistical test is shown in the boxes below:
(a) Is the conclusion from the test in the replication data the same as, or different to, the conclusion that Shalvi et al (2012) reported for the equivalent analysis (see extract above)? (Select one answer):
|
The same conclusion
|
A different conclusion
|
Impossible to say
|
(b) Briefly justify your answer to part (a), making clear what can, or cannot, be concluded from the test for the replication data. (Two or three sentences are sufficient)
(c) Give one reason that might explain why the researchers used a (non-parametric) Mann-Whitney U test to analyse this difference between the two conditions, in preference to using a t-test (as would be more usual when examining a mean difference). (One sentence is sufficient)
[A3: 5 marks available]
Question A4 [12 marks available]
(a) Which of the following research terms accurately describe some aspect of the design or analysis of Experiment 1 of Shalvi et al. (2012):
Select ‘Yes’ or ‘No’ for each term.
|
(i)
|
Experimental research
|
Yes
|
No
|
|
(ii)
|
Quasi-experimental research
|
Yes
|
No
|
|
(iii)
|
Pretest-postest design
|
Yes
|
No
|
|
(iv)
|
Manipulation check
|
Yes
|
No
|
(b) Justify the answers that you gave in part (a) above. (The equivalent of one or two well-written sentences per part (i to iv) will be sufficient for this question.)
[A4: 12 marks available]