Thursday, November 06, 2008

A Review of Willow Creek's Follow Me (Reveal): Answering longitudinal questions with cross-sectional data

(Part 6 in a series)

Turning to concerns about Follow Me, given the similarity of methods of Follow Me and Reveal, they share some of the same weaknesses, and so some of the shortcomings that I discuss here will amplify issues raised in my previous review of Reveal. Follow Me asks an important question: “What moves a person from one stage of spiritual growth to the next” (p. 27). This question is also a longitudinal question in that it concerns how people change over time. What are the sequential stages of spiritual growth? What are the catalysts between these stages? Basically, how and why do people become more spiritually mature over time?

In contrast to the over-time focus of its research questions, the research methods of Follow Me are cross-sectional—portraying each respondent at only one point in time. This disconnection between questions and methods mean causes some problems; specifically, the methods are not well suited for studying change over time. In fact, it’s not clear how much they can at all.

Here’s an illustration. Suppose that we went to a track meet and took a photograph of a race in progress. This photograph would describe the race at one point in time, so it could tell us some things such as who is winning the race and by how much. It would not describe, however, how the race is changing over time. So, just from the photograph, we can not tell if the second-place runner is catching up to the leader or following back. We don’t know how the runners’ relative positions have changed since the start of the race.

Similarly, the analyses of Follow Me tell us which respondents are at which stage at that particular time, but they tell us much less about how the person is changing over time. This means that the study design does not lend itself well to analyses of catalysts and movements and other changes over time. In other words, it's hard to know why some people move from stage to stage while others do not.  Unfortunately, this is the substantive focus of Follow Me.

Follow Me asks an important question, and it uses a perfectly acceptable research design, but it’s the wrong design for the question. The use of cross-sectional data to study longitudinal processes results in various confusion about what’s the cause and what’s the effect, as we’ll see in the next post of this series.

Next: Causal ordering

No comments: