Descriptive statistics

Figure 1. Individual summary statistics
Individual summary statistics

The individual summary statistics seem to indicate that there are three general groups of respondents:

  • Those who will watch the show under a wide variety of circumstances, comprising about 50% of the respondents
  • Those who will watch the show as long as it doesn't change too much, comprising between 40% and 45% of the respondents
  • Those who will watch the show under very strict conditions or not at all, comprising about 10% to 20% of the respondents

Use the Frequencies procedure to obtain a more detailed summary. However, this may be misleading. You cannot be sure from this simple summary that the people who will watch the show because the critics like it are the same people who will watch the show because it's what the people at the office will be talking about. You can gain insight into this situation by looking at the inter-item correlations.

Figure 2. Inter-item correlations
Inter-item correlations

The responses for the first four items are highly correlated, reaffirming the suspicion that people who will watch the show in one of these cases will tend to watch it in other cases. Moreover, there is a positive correlation between each of the first four and last three questions, which was expected. People who are willing to watch the show for more frivolous reasons, such as out of boredom, are likely to watch the show next year if the quality of acting, writing, and directing remains at the level that made the show popular in the first place. On the other hand, people who do not watch the show now are unlikely to watch next year just because they're bored.

While there is a lot of information to be gleaned from looking at correlations, what you really want is a single summary statistic that tells us how reliable our survey is. There are several ways to do this, the most common of which is Cronbach's alpha.

Next