Refer to the Journal of Destination Marketing & Management (October 2015) study of tourist destination image, Exercise 10.11. Recall that the study analyzed the posts submitted on TripAdvisor about The Historic Areas of Istanbul (Turkey) page. In addition to destination image component, the researchers also classified each post according to Type of comment (positive only, negative only, or both positive and negative) and Gender of the poster (male or female). An analysis of the data collected for the sample of 125 tourists who identified their gender is provided in the accompanying Excel/XLSTAT printout.

a. Specify the null and alternative hypotheses for testing whether Type of comment posted to TripAdvisor depends on Gender.

b. Locate the p-value of the chi-square test for independence on the XLSTAT printout. Compare this value with a = .10 and make the appropriate conclusion.

c. Examine the expected (theoretical) cell counts shown on the printout and read the accompanying note at the bottom of the output. Explain why the inference you made in part b may be invalid.

d. Apply Fisher's exact test for the contingency table. Use the exact p-value shown at the bottom of the printout. What do you conclude?

Transcribed Image Text:

Observed frequencies (GENDER / COMMENT): Both Negative Positive 31 61 92 Total Female 15 2. 48 5. Male 11 26 77 Total 125 Theoretical frequencies (GENDER / COMMENT): Negative Positive 2.6880 35.3280 49.0000 4.3120 56.6720 77.0000 92 Both Total Female 9.9840 Male 16.0160 Total 125 26 Test of independence between the rows and the columns (Chi-square): Chesquare (Observed value Chisquare (Critical value) 5.2376 5.9915 DF p-value 0.0729 alpha 0.05 Theoretical counts lower than 5 have been detected. To safely use the Chi-square test based on the Chi-square approximation, the theoretical eounts should not be lower than 5. Fisher's eact test: p-value (Two-taled) 0.0746 alpha 0.05