Page 99 - PhD GT

in the abstract, and Newstead and Collis (1987) state in their discussion that response choices should be chosen and will be understood differently according to the behaviour or attitudes being investigated.
Table4.6
Percentages of responses to each question which are accounted for by the combined responses Rarely and Infrequently and by the response Often
% responses
LDQ item
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10
Rarely / Infrequently Often
8 16 20 12 17 22 19 10 15 14 23 16 13 26 17 26 16 20 16 14
Newstead and Collis (1987) did not find a significant difference between Rarely and Infrequently, though Spector’s data suggest a scale value for Infrequently which is twice the value of that for Rarely (1.7 & 3.8 respectively). One way to reconcile these findings is to follow their recommendation that each population will respond differently depending on the questions they are being asked and it is therefore legitimate to make population and context specific selections of the range of response choices. In the present study, some 20% of the sample said that they did not understand the word Infrequently or that they did not use it. This, therefore did not seem to be a useful response choice to add to this questionnaire. A further question is that regarding the possible inclusion of the response Rarely. Visual examination of responses to the extended LDQ suggested that respondents used either Rarely or Sometimes but not both, and in short interviews a number of respondents reported that they did not use the word Rarely in their everyday thoughts and speech. However, the response Rarely occupied the space between Never and Sometimes which was clear from the rank ordering of the responses shown in Table 4.3.
Another problem thrown up by this study is the question of the difference between the 87