(a) This question required a definition of content analysis which proved challenging for many
students. Almost half of the answers achieved no marks at all. This was made more
remarkable by the fact that most were able to gain some marks on part (b) where they were
asked to explain how to carry out a content analysis for the data in question.
(b) Most students were able to gain some marks here despite poor performance on part (a)
and could identify in a basic way how to carry out a content analysis on the video
recordings. Some were able to provide a clear description of the process but few
appreciated that behavioural categories need to come from somewhere, whether that is
from pilot work or previous research.
(c) Most students were able to identify an appropriate method of testing the reliability of the
content analysis and collect at least one mark. The most popular answers were test-retest
and inter-rater reliability. Many failed to gain the three marks available as their explanation
of how the method of checking reliability would be carried out lacked detail. A few students
became side-tracked into improving reliability and a small number used split half which was
inappropriate in relation to content analysis and gained no marks.
(d) This question required students to explain why a repeated measures design was used in
the experiment. Many students provided a basic answer referring to the need for less
participants or the removal of individual differences but were unable to provide further
explanation of why this would be important in this experiment. Students who thought about
the scenario and elaborated their explanation with reference to reaction times,
concentration or driving skills, achieved full marks.
(e) There was a broad range of answers to this question and about 75% of students achieved
no marks at all. Many students contradicted their previous answer to part (d) and referred
incorrectly to individual differences in reaction times and a similar proportion referred to
order effects which had been controlled by counterbalancing or driving experience. Some
students picked up on the possibility of differences in the nature of the ‘chat’ on the phone
which was encouraging. However, few students showed any awareness of the need to
match the two hazard perception tests (stimulus materials / tasks) in this repeated
(f) Many answers to this question displayed a marked lack of common sense. Despite referring
to a simple hazard perception test, which is a key component of the driving test, many
students claimed that watching a 3-minute film of a road would be traumatic, leading police
drivers to suffer psychological harm. Others referred to possible deception and failed to
appreciate that the purpose of the experiment is rather obvious in a repeated measures
design. Better answers took issues such as informed consent / right to withdraw and
explained how these related to this research.
(g) The question on writing instructions was answered well, with around half of students
achieving four or five marks. Some failed to gain full credit as their instructions referenced
both conditions or failed to include a check of understanding. Very weak answers failed to
refer to the conversation or made no reference to reacting as quickly as possible.
(h) This question required students to identify an appropriate statistical test and justify their
choice. About one third of students gained the full marks for identifying the Wilcoxon test
with appropriate justification but just under half gained one mark only for identification of
the test. Common problems included justification as a test of difference which gained no
credit as it was included in the question. Other students were confused about the type of
Page 154 of 157