Microsoft Office Tutorials and References
In Depth Information
Exhibit 5.15 Extension of row field- resulting table
In order to answer this question, we need not use cross-tabulation analysis.
Cross-tabulation provides insight into how the various characteristics of the respon-
dents relate to preferences; our question is one that is concerned with summary
data for respondents without regard to detailed characteristics. So, let us focus on
how many respondents have a preference for each webpage design? Let’s use the
COUNTIF(range, criteria) cell function to count the number of bad and good
responses that are found in our data table. For Product 1 in Exhibit 5.22, the for-
mula in cell F33 is COUNTIF(F3:F31, “good”) . Thus, the counter will count a cell
value if it corresponds to the criterion that is provided in the cell formula, in this
case good . Note that a split screen is used in this exhibit (hiding most of the data),
but it is obvious that Product 1 , with 18 good , dominates all products. Product 2
and Product 3 are relatively close (15 and 13) to each other, but Product 4 is signif-
icantly different, with only 5 good responses. Again, recall that this result is based
on a relatively small sample size, so we must be careful to understand that if we
require a high degree of assurance about the results, we may want a much larger
sample size than 29 respondents.
Search JabSto ::




Custom Search