Methods of Communication Research and Statistics

A researcher is interested in media-related behaviour among young people and is investigating (among other things) the frequency of reading books, watching television, time spend on the internet, using apps, calling, and number of Facebook friends. She expects to see that young people spend a lot of time on the internet, and that they will have numerous Facebook friends. Database: MediaGedrag.sav.

a. Generate a scatter plot for the 'internet time' and 'Facebook' variables. Using the scatter plot, give your initial impression of the strength and the direction of the association, and check whether there is a linear association.

Strong positive association, yes linear relation

b. Calculate the Pearson's correlation coefficient for the 'internet time' and 'Facebook' variables and provide your interpretation of this.

There is a significant strong and positive association between the time that young people spend on the internet and the number of Facebook friends they have, r = .82, p = .004.

Using the correlation coefficient and the strength and the direction of the relationship, you can conclude that the more time young people spend on the internet, the more Facebook friends they have, and vice versa!

c. Conduct a regression analysis to test the expectation that the more time young people spend on the internet, the more Facebook friends they have. Formulate the null hypothesis for the unstandardised regression coefficient.

In the population, the amount of time spend on the internet is not a significant predictor of the number of Facebook friends.

38.30, someone who does not spend time on the internet has 38.30 Facebook friends

e. What is the value of the unstandardised regression coefficient and what does this mean?

0.57, when time spend on the internet increases by one unit, Facebook friends increase by 0.57 friends

f. What is the value of the standardised regression coefficient and what does this mean?

0.82, very strong positive predictor (when time spend on the internet increases by one standard deviation, the amount of Facebook friends increases by 0.82 standard deviation).

67.5%, the variance of time spend on the internet explains 67.5% of the variance of Facebook friends

h. Is the expectation that young people who spend a lot of time on the internet also have a lot of Facebook friends correct? State all relevant results according to the APA guidelines.

Yes, time spend on the Internet is a significant positive predictor of number of Facebook friends, b* = 0.82, t = 4.08, p = .004, 95% CI [0.25, 0.90].

i. How many Facebook friends does someone who spends 120 minutes a day on the internet have? (calculate this with the regression equation).

SPSS exercise 6.3