Leadership Networks

I have discovered the wonderful Excel tool you have posted for anyone to use but I have a question about the data.

On the SURVEY worksheet, there will be one reply from each respondent. They use a 5-part answer scale from 0 to 4. However, in the example of the COMPILED SURVEYS, the only answer scores shown are 0 and 3. Then on the AUTOMATRIX sheet, 0 and 3 are still the only answers.

If two respondents answered 3 to the same origin-destination “pair”, does it get added together on the AUTOMATRIX sheet?

I would have thought there might be a frequency count for each origin-destination “pair” and each possible answer but that doesn’t appear to be the case.

Example:
0 1 2 3 4
Atkins-Blair : 4 0 15 3 4
Atkins-Baker: 2 2 0 0 9

Can you explain this?

Hi Fran. Thanks for posting your question.

The survey tool assumes that each person answers each question no more than once, so there is no adding to be done. The only possible data about Atkins going to Blair for problem solving is what Atkins said on the survey. It is theoretically possible that we'd ask everybody what they think: "Does Atkins go to Blair for problem solving?" But that is (almost) never done in a network survey. Very different than a "normal" survey!

Furthermore, because the case study data is published in Yes-No form, we have chosen to represent "No" as 0 and "Yes" as 3. If Hansen and Krackhardt had collected and published scalar data, we could have incorporated that into the spreadsheet, but all we have is Yes-No data.

Finally, just in case there is any confusion, note that each row of the spreadsheet indicates the responses of that person. In the example below, Atkins says that he goes to Calder for problem solving. Calder does not say anything about going to Atkins