.
When the participants were selected and agreed to participate in the exercise, a coin was flipped (to ensure randomness) to determine which participant got which questionnaire, whether they suspected their client to be an introvert or extrovert. After completing the questionnaire, choosing a set of questions and stating a short explanation for their choice, a manipulation check ensued. The manipulation check was in the form of a questionnaire also (see Appendix A), asking whether the participant remembered that he/she suspected his/her client to be extraverted or introverted.
After the manipulation check was completed, the participants were thanked and debriefed.
RESULTS: A c test for goodness of fit was used to assess the data from the manipulation check. This was to see whether or not the manipulation of the independent variable, suspicion of introversion or extraversion, was effective and was perceived as expected by the participants. At a=.05 and with a df=1, the obtained value for c=0.45 was not significant (critical c being 3.841). This means that there was no discrepancy between the observed and expected results, the manipulation of the IV was effective. Overall, 40 out of 44 participants correctly remembered the condition given in the questionnaire they received, this amounts to a 90.9% passing rate. Almost all the participants perceived the IV correctly. (For data and calculations see Appendix C.).
For the main results (data and calculations in Appendix D), a c test for independence was used to evaluate the data. At a=.05, df=2 the obtained c=4.25 was not significant, the critical value for c was 5.991. This meant that the participants' choice for the set of questions did not depend on the condition given to them. Therefore, the occurrence of confirmation bias cannot be concluded.
STUDY II:.
METHOD: The participants for the second study were 42 UP Diliman students located in and around the PHAn when the experiment was being conducted.