Business Research: Critical Evaluation Of The Questionnaire - Essay Example

Business Research: Critical Evaluation Of The Questionnaire The questionnaire designed comprised of a mixture eleven questions of varying types and objective. Demographic aspects of the respondents such as gender, age place of origin and postcode of residence were recorded. Caution was taken in designing the questionnaire to ensure that there was a logical flow of the questions and transition of ideas. For instance, the respondents were asked on the amounts they would be willing to pay only after having been asked of their frequency of using the safety bus. Moreover, simplicity of the questionnaire and an easy to read presentation ensured that the questionnaire conformed to the laid out guidelines of a good survey questionnaire[Wal02]. The questionnaire was designed targeting the objectives of the study and all questions were selected in such a way that they obtain information that will be used to answer study questions. More general and simpler questions were put at the beginning to get the respondent into the mood of answering. Further, good questions flow was ensured as it is vital to arrange questions in a way that will not confuse the respondent[Bra04]. In order to make filling of the questionnaire easy and non-confusing for the respondent, clear instructions were included at the start on how to fill the questionnaire. For example the respondent was instructed to tick in the appropriate box that applies to their answer. Questions were kept brief and to the point and as few questions as possible were included since our potential respondents are students who might not be interested or have time to fill long questionnaires. In addition, it was mentioned on the questionnaire the approximate amount of time it will take a respondent to fill the questionnaire[Dör12]. While the questionnaires were to be submitted to respondents by post, the alternative could be the use of online questionnaires. Since the questionnaire was self-administered, there would be no much difference in the response assuming that the student population is a regular internet user and they are computer literate. The proposed postal mode of questionnaire administration may not be as effective as an online questionnaire considering that most students have access to internet and they can just fill in the questionnaire and email it back. This method is much easier than the respondent receiving a posted questionnaire and having to post it after filling. In order to increase response rate, the respondents can be entered into draw to win a prize once they have completed and returned the questionnaires or they can be given a discount ticket for the safety bus for a specific period of time. This will motivate the potential respondents to fill and return the questionnaires. However, the use of incentives in this study should be a point of caution since; this may easily bias responses. Infact, the motivation for student responses might have been the incentives, thus biasing the results. This may impact on the external validity of the analyses results[Lum04]. Data analysis and evaluation The data used for this analysis was collected from 392 students by Tyne and Wear University (TWU) through a questionnaire covering the success of the Safety Bus. The questionnaire was distributed during the summer vacation after the Safety Bus had been operating for one academic year. The sample represented around 11% of the completing first year undergraduate cohort. The variables (responses) recorded for each student was as summarized hereunder[Kir96]. Variable Description Remarks ID Identifier for each student 1 to 392 inclusive. Gender 1=male; 2=female Origin 1=UK; 2=EU; 3=International Postcode Where student lives. Most of these “other” postcode areas are located further from TWU campus than the NX areas. 1=NX1, 2=NX2, 3=NX3, 4=NX4, 5=Other Threatened How often the student feels Threatened or unsafe in the city at night 1=Never; 2=Occasionally; 3=Often; 4=Always Number_Uses Number of uses of the Safety Bus since its introduction Continuous Distance_Home Distance in kilometres) from term-time accommodation to TWU Union Continuous Distance_Public) Distance (in kilometres) from term-time accommodation to nearest public transport, e.g. bus stop Continuous Donation (in pounds) Amounts that student would be willing to pay for a bus ride if they could choose the price they paid. Continuous Rent (in pounds) per calendar month Continuous All statistical tests were performed at 5% significance level and in cases where appropriate, significance at 1% level of significance was evaluated as well[Rog07]. The results of specific analysis are presented in the following sections Respondent profile and feelings of safety As a starting point, graphical summaries and descriptive statistics where appropriate were obtained. The distribution of students in the sample by gender is presented in table 1. There were a slightly higher proportion of females (54.8%) than males (45.2%). The true proportion of males in the population is 48% hence the sample was sufficiently representative of the reality. This therefore guarantees that the results can be generalized to the population of interest, thus giving the results external validity[Tim05]. Table 1: Distribution of respondents by gender   Frequency Percent Male 177 45.2 Female 215 54.8 Total 392 100 A pie chart of the sampled students based on their origin (figure 1) revealed that the largest proportion of students (79.34%) were from the UK, while only 7.40% were from the EU. The rest were international students. Similarly, evaluation of the distribution of students based on their postcode was done. Table 2 presents the results from which it was observed that only 10.5% of the students were from postcode areas located further from TWU campus than the NX areas. Table 2: Distribution by postcode   Frequency Percent NX1 44 11.2 NX2 161 41.1 NX3 46 11.7 NX4 100 25.5 Other 41 10.5 Total 392 100 Approximately 40% of the students occasionally felt threatened or unsafe as compared to 13% of those that always felt threatened. Moreover, 14% of the students never felt threated at all (figure 2). To further determine the association between gender and perception towards being threatened or unsafe, cross-tabulations (table 3a) as well as Pearson’s chi square test for association (table 3b) was performed. From the cross-tabulation, there was some evidence of variation of perception on safety with gender, with more females (39) feeling always threatened as compared to males (12). This was further confirmed by the Pearson’s test which was significant (P-value=0.0001)[Kut05]. Table 3 a: Cross-tabulation of threatened by gender     Never Occasionally Often Always Total Gender Male 31 85 49 12 177 Female 24 69 83 39 215 Total   55 154 132 51 392 Table 3b: Pearson Chi-Square test   Value df Asymp. Sig. (2-sided) Pearson Chi-Square 22.1292 3 0.0001 The perception of being threatened was evidently different for students based on their origin. Caution should be exercised however, since the large disparities may be as a result of varying sample sizes by region. For instance the UK had 311 students in the sample as compared with International students (Table 4). A statistical test however provided no evidence of an association between the place of origin of the student and the feeling of being threated (P-value=0.296785). Table 4: Cross tabulation of threatened by Origin     Never Occasionally Often Always Total Origin UK 45 121 107 38 311 EU 4 10 13 2 29 International 6 23 12 11 52 Total   55 154 132 51 392 Table 5 presents the results of perception towards threats according to the student’s postcodes. There was quite a disparity in the students’ responses on insecurity based on their postcodes as was evident from the Pearson’s Chi square test P-value of 0.00037[Men07]. Table 5: Threatened by Postcode     Never Occasionally Often Always Total Postcode NX1 8 21 13 2 44 NX2 31 68 48 14 161 NX3 4 14 20 8 46 NX4 3 42 33 22 100 Other 9 9 18 5 41 Total   55 154 132 51 392 Characteristics of Safety Bus users On average, the sampled students had used the safety buses for approximately three times since its introduction. This is evident from the mean, mode and median (50th percentile) which are all about three (Table 6). Table 6: Number of Uses     Mean 2.96 Mode   3 Percentiles 25 2 50 3   75 4 To confirm whether the hypothesized number of uses of the safety bus (2) was valid, a t-test was conducted (Table 7). There was a significant difference in the number of uses of the safety bus. Actually most students used the bus three times on average (Mode=3). Table 7: T-test for the hypothesized number of uses   t df Sig. (2-tailed) Mean Difference 95% Confidence Interval of the Difference       Lower Upper No_Uses 10.269 391 0 0.957 0.77 1.14 Further analysis involved checking whether the amount of safety bus usage varied with gender, origin or postcode. There was no significant difference in usage of the safety bus between males and females (From table 8, the P-value was 0.521). Table 8: Test for difference in usage by gender   Sum of Squares df Mean Square F Sig. Between Groups 1.404 1 1.404 0.412 0.521 Within Groups 1328.858 390 3.407 Total 1330.263 391       Similarly, the safety bus usage amongst students in the Tyne and Wear University (TWU) was not significantly different across the regions of origin UK, EU or International (Table 8; P-value=0.542). Table 9: ANOVA by Origin   Sum of Squares df Mean Square F Sig. Between Groups 4.185 2 2.092 0.614 0.542 Within Groups 1326.078 389 3.409 Total 1330.263 391 The postcode from which a student resided was however, found to have a significant influence on the number of uses of the safety bus (Table 10). Further tests can be conducted to establish how the mean usage varied between pairs of postcodes. Table 11: ANOVA by Postcode   Sum of Squares df Mean Square F Sig. Between Groups 153.969 4 38.492 12.664 0 Within Groups 1176.294 387 3.04 Total 1330.263 391       Predicting price willing to pay for the Safety Bus There was also interest in establishing the amount students would be willing to pay for a journey. In particular, there was need to compare the average donations students were willing to make, with the standard current fixed payment of £1. On average, students would be willing to contribute £1.04 for their journey on the safety bus. Both the 95% and 99% confidence intervals included £1. Thence we conclude that the amount of donations students would make was not significantly difference from the fixed value (Table 12). Table 12: Confidence intervals for the mean donation Statistic Std. Error Lower Bound Upper Bound 95% Confidence Interval for Mean 1.04 0.0316 0.978 1.102  99% Confidence Interval for Mean 0.9583 1.122 A boxplot of the donations students would be willing to make revealed that there was no much variability from the fixed value of £1. There was however quite a number of students willing to make more donations with the most outlying one (ID=239) contributing £4[Tim05]. Correlation test was performed on the continuous variables donation, distance from term-time accommodation to TWU Union, distance from term-time accommodation to nearest public transport and the amount of rent per calendar month. The distance from term-time accommodation to TWU Union had the highest correlation (0.95) with the amount of donations, while there was very little correlation (0.099) between donations and rent paid per calendar month. Table 13: Correlation Martix   Donation (£) Distance_Home (km) Distance_Public (km) Rent (£) Donation (£) Pearson Correlation 1 .950(**) .634(**) -.099(*) Sig. (2-tailed) 0 0 0.05 N 392 392 392 392 Distance_Home (km) Pearson Correlation .950(**) 1 .674(**) -0.093 Sig. (2-tailed) 0 0 0.067 N 392 392 392 392 Distance_Public (km) Pearson Correlation .634(**) .674(**) 1 0.015 Sig. (2-tailed) 0 0 0.773 N 392 392 392 392 Rent (£) Pearson Correlation -.099(*) -0.093 0.015 1 Sig. (2-tailed) 0.05 0.067 0.773 N 392 392 392 392 ** Correlation is significant at the 0.01 level (2-tailed).   * Correlation is significant at the 0.05 level (2-tailed).   The correlations were better visualized in a scatter plot matrix as shown below. Consistent with the previous results, there was a strong positive correlation between donations and the distance from term-time accommodation to TWU Union. This correlation implies that distance_home can explain 95% of variability in donations if used in fitting a regression model alone. Therefore, a regression model with distance_home as a covariate was fit and the results presented in table 14. Distance_home was significant in explaining the mean variation of donations. The model had an R square value of 0.9026 implying that it could explain 90.26% of variability in the donations[Kut05]. Table 14: Regression model Model Unstandardized Coefficients t Sig.     B Std. Error Std. Error (Constant) 0.089 0.019 4.798 0   Distance_Home (km) 0.2 0.003 60.123 0 The resulting regression model can be denoted as . Predictions can now be done based on this model for the given students. Their predicted donations are as tabulated hereunder. Distance_Home (km) Formula Predicted donation Student 1 7.5 0.089+0.2(7.5) 1.589 Student 2 25 0.089+0.2(25) 5.089 Discussion and conclusions Approximately 85% of the students had some level of anxiety regarding their safety, while 13% of the students always felt threatened. Moreover, females felt insecure more than male students. With the introduction of the safety bus, students surpassed the hypothesized usage of the bus of two times, with most of them using the safety bus 3 times since its introduction. The amounts the students were willing to donate in a new pricing system were not significantly different from the current fixed price of £1. One of the implications of the findings of this report is that there might be no need to change the current billing system since; in terms of revenue, there wouldn’t be much difference. Moreover, considering the high uptake of the safety bus service, the management should monitor the need for additional safety buses. This survey however, did not address the issues of student satisfaction with the service. Future studies therefore, should take into account user satisfaction in the questionnaire design and analysis. References Wal02: , (Walden, 2002), Bra04: , (Brace, 2004), Dör12: , (Dörnyei & Taguchi, 2012), Lum04: , (Lumley, 2004), Kir96: , (Kirkman, 1996), Rog07: , (Rogelberg & Landis, 2007), Tim05: , (Urdan, 2005), Kut05: , (Kutner, et al., 2005), Men07: , (Mendenhall & Sinchich , 2007), ID Number Tyne and Wear University (TWU) safety bus study Please take a few moments to complete our safety bus satisfaction survey. Your responses will help us to address any issues that you may have as well as to better target our services to meet your needs. Your responses will be kept confidential. This survey will take approximately 5 minutes to complete. Tick where appropriate 1. Date of birth __dd__/__mm__/__yyyy__ 2. What is your Gender Male Female 3. What is your Origin United kingdom European Union International 4. What is your Postcode _____ 5. Programme enrolled in________________________ 6. How many times have you used the Safety Bus since its introduction _________ 7. What is the distance (in kilometres) from TWU Union to your term-time accommodation? ____________ 8. What is the distance (in kilometres) from term-time accommodation to nearest public transport, e.g. bus stop? ____________ 9. How much (in pounds) would you be willing to pay for a bus ride if you could choose the price to pay? __________ 10. How often do you feel Threatened or unsafe in the city at night Never Occasionally Often Always 11. What is your Rent (in pounds) per calendar month? ___________ Coding plan Gender 1=Male 2=Female Origin 1=United Kingdom 2=European Union 3=International Postcode 1=NX1 2=NX2 3=NX3 4=NX4 5=Other Threatened 1=Never 2=Occasionally 3=Often 4=Always Read More