Cluster Analysis Overview

The following includes an overview of the cluster analysis conducted as a component of the OVBC 2023 Typology Study. The information below includes some basic methodology, as well as how to read, interpret, and understand the tables that display the eight clusters’ responses to the 21 A/B statements used in cluster analysis. The tables are included below, and are available to download as an attachment from the bottom of the page.

Study Points to Common Ground Among Oregonians

A two-month study conducted by the independent, non-partisan Oregon Values and Beliefs Center (OVBC) concludes that regardless of political orientation, there is common ground among Oregonians on key public policy issues (e.g. government efficacy and efficiency, environmental regulations, etc.)

These findings point to a variety of issues that bridge gaps often thought to define Oregonians, such as: urban vs. rural; Republican vs. Democrat; liberal vs. conservative; Portland vs. the rest of the state; etc.

Building on past research conducted by the nation Pew Research Center and Oregon’s Policy Interactive, OVBC developed 21 pairs of ideological statements about government, economics, the environment, and social issues to gauge respondents’ opinions. Each pair included a left-leaning and right-leaning statement. (See the red and blue Typology Statements: Score by Cluster table below)

Data analysis, which included statistical weighting of key demographics to ensure an Oregon-representative sample and cluster analysis determined eight categories that best distinguish Oregonians age 18+. The categories, which are based on typology statement preferences, are:

  • Cluster 1: Party-Aligned Progressives
  • Cluster 2: Dispassionate Liberals
  • Cluster 3: Alienated Young Left
  • Cluster 4: Green Rural Independents
  • Cluster 5: Diverse and Devout
  • Cluster 6: Disengaged Traditional Conservatives
  • Cluster 7: Free-Market Libertarians
  • Cluster 8: Modern Conservative Loyalists

The results show that Party-Aligned Progressives and Modern Conservative Loyalists each represent 23% and 9%, respectively, of the sample, with the other 6 archetypes or “neighborhoods” populating the remaining 68%. Rather than a linear spectrum of the left (blue) to right (red), the A/B Statement table reveals a graphic checker-board matrix of blue, white, and red in which the categories leaning more conservative or more liberal fluctuate from question to question.

In addition to the 21 pairs of left-leaning and right-leaning statements used for the development of the typology clusters, the full study included questions about specific statewide planning, policy, communications issues, and demographics. Aligning the findings from these questions with the cluster populations provides fresh insights into the values and beliefs of Oregonians. (See the yellow and green highlighted Demographic or Survey Questions table below)

Future reporting will include cluster demographic and attitudinal profiles. These documents will show where there’s strong and moderate agreement among Oregonians on key public policy issues, as well as areas of contention.

Reading the Typology Statements: Score by Cluster Table

Over 3,000 respondents completed the 60-question survey used for the 2023 Oregon Values and Beliefs Typology Study. Statistical software was utilized to weight a subsample (n=2,610 Oregonians) by five demographic characteristics to ensure a representative sample of the state’s population age 18+ and to complete a cluster analysis of the responses to the 21 typology questions included in the full questionnaire. Different numbers and combinations of clusters were compared and considered, and an iteration using eight clusters was selected by the researchers as being the most definitive of our values and beliefs. The questions used for categorization and the corresponding responses of each cluster are detailed in the table on the following page.

The Typology Statements: Score by Cluster table below has a row for each of the 21 A/B statement typology questions, and 9 columns with one column for each cluster and an additional column showing the overall mean score for that question. Below is the first of the 21 typology questions.

The cells were scored using a four-point liberal to conservative forced-choice scale. To accommodate comparative analysis with similar Pew Research studies at the national level, the scale was then converted to an equivalent 7-point liberal to conservative scale ranging form -3 (liberal, strong), 0, +3 (conservative, strong). For this question, the far left cell shows a moderately-liberal score of -1.3 out of a maximum -3, while the equivalent right side exhibits a strong +2.6 conservative score out of a possible +3. The color coding is from left “blue” (negative/liberal score) to right “red” (positive/conservative score). The darker shading indicates higher scores.

The Demographic or Survey Questions table further below shows cross-tabulations of selected survey questions aligned with the cluster columns: Below is the first row for gender.

The first cell contains the demographic, or question. The next 8 cells, left to right, indicate the sample population percentages for each cluster column. Therefore, 61% in the 2nd cell under C1 means a 61% population of women in Party-Aligned Progressives. Likewise, the far-right cell under C8 indicates a make-up of 37% of women in Modern Conservative Loyalists. This is consistent with conventional research which has found women are more prone to liberal values and men more toward conservative values.

The following tables serve as snapshots within a wealth of data, and an effort to encapsulate as much information as possible from the broader picture.



An introduction and overview to the OVBC 2023 Typology Study Cluster Analysis, including how to read and interpret the cluster analysis tables