To determine whether there was any difference in the responses among the two waves of respondents, we performed independent samples t-tests for the 19 scale items and Chi-Square tests for the 7 multiple choice and demographics questions. There were significant differences for three of the nineteen scale items: trust 4( t284 = −2.613,p=.009), friendship 2( t285 = −2.12, p=.035), and friendship 4( t285 = −1.98, p=.049). To determine whether this nonresponse bias is a significant threatt o our results,we calculated the group differences for the respondents from the two waves. The results in Table VIII show that the scale means were higher for the second wave respondents for all four constructs. Therefore, because the second wave of respondents had higher means (significanly higher in two cases) than the first wave of respondents, the non-response bias is not a serious threat in this study.
To determine whether there was any difference in the responses among the two waves of respondents, we performed independent samples t-tests for the 19 scale items and Chi-Square tests for the 7 multiple choice and demographics questions. There were significant differences for three of the nineteen scale items: trust 4( t284 = −2.613,p=.009), friendship 2( t285 = −2.12, p=.035), and friendship 4( t285 = −1.98, p=.049). To determine whether this nonresponse bias is a significant threatt o our results,we calculated the group differences for the respondents from the two waves. The results in Table VIII show that the scale means were higher for the second wave respondents for all four constructs. Therefore, because the second wave of respondents had higher means (significanly higher in two cases) than the first wave of respondents, the non-response bias is not a serious threat in this study.
การแปล กรุณารอสักครู่..
