Hi, please can someone help me out. I would like to know if I have performed the correct analysis for my study. I am comparing direct response options to likert scale items. I will paste my analysis below, any suggestions or assistance would be much appreciated. Thank you all for reading this and for your time.

4.5 Data Analysis
The data from this study was analysed using IBM SPSSTM Version 23, statistical software as it was easy to use, economical and contains the necessary statistical functions for analysis of the data obtained. An summary of the analyses performed for this study can be seen in the table below.
Table 8: Summary of Data analysis

Question Method of data entry Data analysis
Criterion 1 (sensitive/ non-sensitive) Counts obtained by tallying responses Counts, mean scores, standard deviations and reliability coefficients
Criterion 2 (True/ False) Counts obtained by tallying responses Counts, mean scores, standard deviations and reliability coefficients
Norming Option Extreme scores combined in order to produce dichotomous data (agree/ disagree)for comparison with Criterion question, factor loadings obtained with Likert scale items Factor analysis, Counts mean scores, standard deviations and reliability coefficients
Social Desirability Scale Extreme scores combined in order to produce dichotomous data (agree/ disagree) for comparison with Criterion question Counts, mean scores, standard deviations and reliability coefficients

In order to achieve the aims of this study, perceptions on sensitivity and disclosure were analysed by tallying the responses obtained using the Count option on SPSS. This was done as perceptions of sensitivity and disclosure made use of direct response options, therefore participants were explicitly indicating if the behaviour in question was sensitive or not and/or if the behaviour was true or false for them. In order to ensure that the correct conclusions were drawn, any behaviour with a 40% endorsement or more (126 counts or more) was regarded as an indication of sensitivity and/or disclosure of the behaviour in question. Mean scores, standard deviations and reliability coefficients were also carried out as a means of assessing the two response options.
In terms of the norming option, factor analysis was carried out using principal components analysis. Since the aim of this study was to norm and scale sensitivity, it was essential to ensure that the question posed was able to measure sensitivity. Items that correlated .4 or higher were regarded as sensitive. The .4 factor loading was chosen as it represents the bulk of the data. To enable a comparative analysis between the direct response options, namely sensitivity and disclosure, the responses of the norming option were dichotomised by combining the two extremes. This provided an agree and disagree response format which was tallied and compared to the direct response options.
The social desirability scale utilized in the norming study, was analysed by calculating a reliability coefficient. Cronbachs alpha was calculated as part of establishing the reliability of the scale utilized. Mean scores were then calculated and the two extreme scores of the data was dichotomised in order to achieve a social desirability score. This enabled or possibly eliminated wrongful classifications, in terms of obtaining marginal classifications for social desirability (Hays et al, 1989).


Please let me know what you think.