Talk:Multistage sampling

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Marketing & Advertising Mid‑importance This article is within the scope of WikiProject Marketing & Advertising, a collaborative effort to improve the coverage of Marketing on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Marketing & AdvertisingWikipedia:WikiProject Marketing & AdvertisingTemplate:WikiProject Marketing & AdvertisingMarketing & Advertising Mid This article has been rated as Mid-importance on the importance scale. Measurement (defunct) This article is within the scope of WikiProject Measurement, a project which is currently considered to be defunct.MeasurementWikipedia:WikiProject MeasurementTemplate:WikiProject MeasurementMeasurement Statistics Mid‑importance This article is within the scope of WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.StatisticsWikipedia:WikiProject StatisticsTemplate:WikiProject StatisticsStatistics Mid This article has been rated as Mid-importance on the importance scale.

Probability-proportional-to-size sampling is a type of multistage cluster sampling. In this method, the probability of selecting an element in any given cluster varies inversely with the size of the cluster.

I removed this bit because (a) PPS is not necessarily a multistage technique (it's used as a single-stage technique in e.g. samples of businesses), and (b) selection probability in PPS varies proportionally to size, not inversely.

I suspect this might come from some confusion about multi-stage designs where the ultimate object is to give all bottom-level elements the same selection probability. One way to do this is by selecting clusters PPS in the first stage (bigger clusters have a higher probability of selection) and then applying inverse of cluster size in the second stage, when we're sub-selecting within clusters. But in this case, the 'PPS' bit is the first stage, and the 'varies inversely with size' bit is the second stage - they're not the same thing. --144.53.226.17 (talk) 03:56, 5 January 2009 (UTC)Reply

This article is as bad as cluster sampling (which mixes up stratified sampling with cluster sampling). Multistage sampling should be renamed Complex sampling cf. current literature, most of the contents should be moved to cluster sampling (which should briefly cover both two-stage and one-stage cluster sampling, and briefly with and without replacement). Complex sampling should then cover multistage, weighting, stratification and post-stratification all at once (but each of them can have their own pages). That would be very clear and good for those wanting to learn. I don't mind doing it with enough votes. Sda030 (talk) 05:23, 26 February 2014 (UTC)Reply