Skip to main content

Stopping Rules for Surveys

One aspect of responsive design that hasn't really been considered is when to stop collecting data. Groves and Heeringa (2006) argue that you should change your data collection strategy when it ceases to bring in interviews that change your estimate. But when should you stop?

It seems like the answer to the question should be driven by some estimate of the risk of nonresponse bias. But given that the response rate appears to be a poor proxy measure for this risk, what should we do?

Rao, Glickman and Glynn proposed a rule for binary survey outcome variables. Now, Raghu and I have an article accepted at Statistics in Medicine that proposes a rule that uses imputation methods and recommends stopping data collection when the probability that additional data (i.e. more interviews) will change your estimate is sufficiently small. The rule is for normally distributed data. The rule is discussed in my dissertation as well.
Labels:

Comments

  1. Andy PeytchevDecember 1, 2009 at 12:32 PM

    James, I really like the idea of stopping rules and yours seems very intriguing. I hope to read your article soon :) Without having read it, it makes me wonder how estimates for different variables and different statistics (e.g., MSE) can be handled in any one of these approaches. Within a responsive design, however, it seems even more difficult to incorporate cost. When do you release a few more sample replicates instead of collecting increasingly expensive interviews? Not only does it bring in cost, but such a stopping rule would incorporate variance that is not due to nonresponse. Mindboggling, but thanks for the thought-provoking post.

    ReplyDelete
  2. Andy PeytchevDecember 1, 2009 at 12:59 PM

    For some reason my comment did not get posted... James, I wonder how contradictory criteria can be managed, such as MSE. The other somewhat related part that is missing from stopping rules are tradeoffs, such as cost (I look forward to reading your article - this is just a thought in general). At what point do you decide to add additional sample replicates instead of pursuing expensive cases, even if they still can affect estimates? Both stopping and releasing replicates are options that may contradict each other, or rather, nonorthogonal options. It would go beyond nonresponse bias and variance...

    ReplyDelete
  3. James WagnerDecember 2, 2009 at 9:38 AM

    Thanks for the comments! The criteria are not necessarily contradictory. The rule does depend on a sufficient sample size to determine that you are satisfied that the risk of your estimate changing with further data collection is small. What the sample size might be, you won't know ahead of time (you might try simulations).

    Of course, surveys also have targets for sampling error, so I suggest that you have another rule for that target. Once both rules have been met, you can stop.

    I haven't incorporated cost. In simulation, it looks like we might be going too long on some surveys. But you could imagine we are stopping too soon on others.

    There's a lot more to do. Currently, this setup relies on a single statistic (as does most sample design theory). How do we handle multi-purpose surveys? As you mention, cost issues are another issue.

    ReplyDelete
  4. dewiqq.ccJuly 18, 2019 at 11:51 AM

    Thanks for sharing quality
    will help you more:
    customer survey

    ReplyDelete

Post a Comment

Popular posts from this blog

Tailoring vs. Targeting

One of the chapters in a recent book on surveying hard-to-reach populations looks at "targeting and tailoring" survey designs. The chapter references this paper on the use of the terms among those who design health communication. I thought the article was an interesting one. They start by saying that "one way to classify message strategies like tailoring is by the level of specificity with which characteristics of the target audience are reflected in the the communication." That made sense. There is likely a continuum of specificity ranging from complete non-differentiation across units to nearly individualized. But then the authors break that continuum and try to define a "fundamental" difference between tailoring and targeting. They say targeting is for some subgroup while tailoring is to the characteristics of the individual. That sounds good, but at least for surveys, I'm not sure the distinction holds. In survey design, what would constitute
36 comments
Read more

What is Data Quality, and How to Enhance it in Research

  We often talk about “data quality” or “data integrity” when we are discussing the collection or analysis of one type of data or another. Yet, the definition of these terms might be unclear, or they may vary across different contexts. In any event, the terms are somewhat abstract -- which can make it difficult, in practice, to improve. That is, we need to know what we are describing with those terms, before we can improve them. Over the last two years, we have been developing a course on   Total Data Quality , soon to be available on Coursera. We start from an error classification scheme adopted by survey methodology many years ago. Known as the “Total Survey Error” perspective, it focuses on the classification of errors into measurement and representation dimensions. One goal of our course is to expand this classification scheme from survey data to other types of data. The figure shows the classification scheme as we have modified it to include both survey data and organic forms of d
Post a Comment
Read more

An Experimental Adaptive Contact Strategy

I'm running an experiment on contact methods in a telephone survey. I'm going to present the results of the experiment at the FCSM conference in November. Here's the basic idea. Multi-level models are fit daily with the household being a grouping factor. The models provide household-specific estimates of the probability of contact for each of four call windows. The predictor variables in this model are the geographic context variables available for an RDD sample. Let $\mathbf{X_{ij}}$ denote a $k_j \times 1$ vector of demographic variables for the $i^{th}$ person and $j^{th}$ call. The data records are calls. There may be zero, one, or multiple calls to household in each window. The outcome variable is an indicator for whether contact was achieved on the call. This contact indicator is denoted $R_{ijl}$ for the $i^{th}$ person on the $j^{th}$ call to the $l^{th}$ window. Then for each of the four call windows denoted $l$, a separate model is fit where each household is assu
4 comments
Read more