Cluster Analysis Crosstables

The Cluster Analysis Crosstables include each question asked in the OVBC 2023 Typology Study, and all possible response categories. These crosstables also include columns for each of the eight clusters or “neighborhoods” of shared values and beliefs. For example, the excerpt below examines what percentage of each cluster, 1-8, selected each possible response to Q2, right alongside the percentages for the full sample.

The full Cluster Analysis Crosstables can be downloaded from the bottom of this page.

We would like to acknowledge and thank our research partner, PolicyInteractive.

HOW TO READ CROSSTABLES


In the example chart (left), rows represent the five possible response options to the question “All things considered, do you think your state is headed in the right direction, or is it off on the wrong track?”

The five response options are: “Right Direction, strongly,” “Right Direction, somewhat,” “Wrong Track, somewhat,” “Wrong Track, strongly,” and “Don’t know/Not sure.” [If responses are italicized, the answer choices that were given to a respondent have been collapsed together. (i.e. If we wanted to see all people who said “Right direction” by collapsing the options “Right direction, strongly” and “Right direction, somewhat” there may be an italicized “Right direction” option as well)]

The last column, or banner, is the “Total” column. It reflects the overall results, or the total amount and percentages in each response category.

The other eight columns reflect the results from the subgroups of the typology study. The gender subgroup is composed of Men, Women, and Non-binary or other gender identification, each with their own column. Each subgroup is assigned a letter, for example: Men=A Woman=B, Non-binary=C.

Any sampling of opinions or attitudes is subject to a margin of error, which represents the difference between a sample of a given population and the total population. The margin of error is a statistic expressing the amount of random sampling error in a survey’s results and differs by sample size, as reflected in the table. Cluster analysis is subjective, as there is no universal way to cluster data. To account for this limitation, multiple iterations of cluster analysis were conducted by OVBC and PolicyInteractive researchers before this set of clusters was chosen. Many of the iterations had similar groupings or clusters of demographics, values, and beliefs, but these eight were the most clearly defined and collectively best describe Oregonians 18+.