Surveys are a unique data source, both in terms of how the data is collected and how the data is stored. Standard analysis techniques such as linear modeling can lead to misleading results if naively applied to stratified or clustered survey samples. Important metadata, often attached in a header or companion file, can be lost if the wrong import steps are used. This session shows JMP 12 features designed to make survey analysis easier, as well as advanced JMP techniques that can be used to analyze survey data in a way that respects sampling design. These techniques can allow researchers to discover new patterns in their survey studies, as well as speed up standard analyses using unique features in JMP. The methods shown can be applied to any data set with a large number of categorical variables.
Instructions for running the code examples:
JMP cannot distribute the survey results, so you need to download the data from the NEA website and run the data preparation scripts to get started. Use the following steps: