In recent years, many activities have to some extent transitioned to the internet. One of them is filling out a survey questionnaire, which has started to face new challenges of its new environment, such as non-response. In my thesis, I explored the effect of separative questions and horizontal scroll matrix on non-response in case of large scales in a web survey. In the hypotheses, I predicted that the use of separative questions and horizontal scroll matrix would have a lower rate of non-response compared to non-use. In using separative questions and horizontal scroll matrix, I was not able to determine a statistically significant effect on lower response rate of the respondents who completed the survey using a computer. A statistically significant impact on lower non-response rate was found only in the use of separative questions in the respondents who completed a telephone survey. I also found that the distribution of the table has a statistically significant effect on lower response rate of the respondents who completed the survey using a computer. Compared to the method without the distribution of the table, the use of separative questions and horizontal scroll matrix, other methods had a statistically significant effect on lower non-response rate. The results showed that it is advisable to divide the table with large scales into several smaller sets and to use separative questions, while the use of horizontal scroll matrix does not have a significant impact on non-response.
|