DEPARTMENT OF PSYCHOLOGICAL SCIENCES
RESEARCH METHODS 2
CREDIT VALUE: 15 credits
DATE OF EXAMINATION: 3 September 2020
This exam has three sections A, B, and C. Make sure to look at all sections.
(Answer ALL questions)
QA1 (36 marks)
Give a brief description of each of the following, illustrating your answers with
examples where appropriate.
a) Post-hoc contrasts in ANOVA (12 marks).
b) The ways in which multiple predictors can be entered into a multiple linear
regression and how they differ (12 marks).
c) The potential types of participant harm that would be avoided in an ethically
approved study. (12 marks).
(Answer ALL questions)
QB1. (50 marks).
A cognitive psychologist is interested in the effects of emotional facial expressions on
spatial visual attention. She believes that happy facial expressions capture attention
in real-world social situations, while people look away from sad expressions. She
conducted an experiment using the classic ‘Posner cueing’ paradigm. In this
paradigm, participants have to respond to a target stimulus that appears on either
the left or the right side of the screen by pressing one of two buttons to indicate
which side of the screen it appeared on. Before the target appears, a task-irrelevant
cue appears either in the same location or the opposite location. In a classic paper,
Michael Posner and colleagues showed that reaction times are faster when the cue
appears on the same side as the target (congruent trials) than when the cue appears
on the opposite side (incongruent trials). In the present study, the psychologist
predicted that using a happy facial expression as the cue will cause participants to
respond faster than using sad or neutral facial expressions. She also predicted that
the effect of emotional expression should be larger for congruent than for
incongruent trials. As stimuli, she used line drawings of faces with different emotional
expressions. The overall structure of the experiment is shown in Figure 1.
Figure 1: Experimental timeline. Each trial started with a fixation cross in the centre
of the screen. Then, a cue face appeared (either happy, neutral, or sad) on either the
left or right size of the screen. Finally, a target stimulus appeared either on the same
side as the cue (congruent trial) or on the opposite side (incongruent trial).
Participants completed 20 trials of each combination of the three types of facial
expression and the two types of cue-target congruency. Trials in which participants
responded incorrectly were excluded from analysis. The mean reaction time (in
milliseconds) for each participant in each of these conditions is presented in Table
B1. The relevant SPSS output is presented in Table B2.
1. What type of ANOVA should be conducted on these data? (3 marks)
2. How many factors are there and how many levels does each have? Please
provide a description of the levels in each factor (4 marks)
3. Report the main effects and interactions using the traditional notation and
describe if they are significant or not (12 marks)
4. What main effects and/or interactions did the researcher predict and are they
supported by the results? (6 marks)
5. Using the means displayed in Table B1, identify which of the four plots below
correctly represents the two intervention groups across the original BMI. Using
the graph, explain what is likely to be driving the effects reported in. (10
1. Describe the two ways that follow-up analyses could be performed to check
where the effect lies? (6 marks)
2. Explain three issues and potential improvements that may be made to the
study (9 marks)
Table B1: Mean reaction time (milliseconds) for responses to target stimuli in the
two congruency conditions and the three emotional expression conditions.
Table B2: SPSS output
(Answer ALL questions)
QC. (19 marks)
A team of researchers were interested in how background noise impacts cognitive
performance. To investigate this question they split adult participants into three
groups and asked them to perform a difficult maths task whilst either in a silent room
(group 1), listening to one person talking over headphones (group 2) or the sound of
a busy crowded café (group 3). Researchers collected accuracy on the task. They
predicted that the one person talking condition (2) would be more distracting than
the other two conditions and show more negative impact on performance.
1. What type of trend (i.e. shape of the effect) would the researchers have to
examine to test their hypothesis? (1 mark)
2. Using the traditional notation, report whether the analysis presented in table
C2 indicates if there was a significant main effect of Audio Groups and what
the shape of the effect was, i.e. the trend. (8 marks).
3. With reference to the descriptives in Table C1 which graph corresponds to
participant results and why (3 marks)?
1. Should the researchers conclude that their hypothesis was supported? (2 marks)