- Is there a significant relationship between time spent online and gender? (Cross Tabulation)
- What is the correlation between age and likeliness to recommend Dell products to friends and family? (Correlation Test)
- Does a degree of how likely a consumer is to recommend Dell, have a significant difference in their evaluations of their own qualities of opinion leadership in terms of technology? (Hypothesis Test for Differences T-Test)
- Can the likelihood of recommending Dell be explained in terms of ratings consumers have for the various attributes of Dell computers on an individual basis? (Regression Test)
Research Question 1:
- Null:There is a significant relationship between the time spent online and gender.
- Alternate: There is not a significant relationship between time spent online and gender.
Research Question 2:
- Null: There is a strong correlations between age and likeliness to recommend Dell products to friends and family.
- Alternate: There is not a strong correlation between age and the likeliness to recommend Dell products to friends and family.
Research Question 3:
- Null: Consumer's likelihood of recommending Dell changes significantly based on consumer's evaluations of their own opinion leadership.
- Alternate: Consumer's likelihood of recommending Dell does not change significantly based on consumer's evaluation of their own levels of opinion leadership.
Research Question 4:
- Null: Individual's likelihood of recommending Dell is strongly tied to their ratings on the various attributes of Dell computers.
- Alternate: Individual's likelihood of recommending Dell is not strongly tied to their ratings on the various attributes of Dell computers.
Cross Tabulation: All of the Chi values were measured at less than .001, which concludes the chance of error very low. The cross tabulation indicates that men are more closely distributed between light, medium, and heavy internet users. More than half the women fall within the light users category. This indicates that there is a some significant relationship between gender and the time spent online. However, the strength of the correlation is weak with the Cramer's V at .217, which means a low association.
Correlation Test: The Pearson correlation for age and likelihood of recommendation is very weak at .004. There is not a strong correlation between age and likeliness of recommendation. We would accept the null hypothesis of there not being a significant correlation.
Hypothesis Test for Differences (T-Test): All three tested opinion leaderships variables did not signify a significant difference between the group means. Since each coefficient was greater than 0.05, there is no significant difference between recommendation and opinion leadership attributes.
Regression Test: The linear regression test indicated that the questions of: "If you could make your computer purchase decision again, how likely would you be to choose Dell?", "How much do you agree that Dell computers has high quality computers with no technical problems?", and "How much do you agree the Dell computers has high quality peripherals?" are all significant. These questions were tested as less than 0.05, which means that they do have a high level of predictability.
Cross Tabulation: Dell should recognize the low association relationship between gender and time spent online. Therefore, if Dell were trying to market to women, the computer's capability for online activity would be less valuable than if Dell were marketing to men.
Correlation Test: In order to find opinion leaders, Dell must look for characteristics outside age. Age is relatively unimportant to whether or not an individual considered themselves to be an opinion leader.
Hypothesis Test for Differences (T-Test): Dell should not rely on the opinion leadership qualities of the consumer to determine whether the consumer would or would not recommend the product.
Regression Test: Dell should focus on analyzing the consumer's choice which involves technical problems and high quality peripherals. This will help ensure that Dell is experiencing high levels of satisfaction and a high likeliness of repeat purchases and recommendations.
Because there were insignificant conclusions, the research could be conducted focusing on attributes other than age. There could also be different focuses when it comes to recommendation, since the results for opinion leadership were also found insignificant.