Why We Did This
The Oregon Voices survey has been many years in the making. We wanted a research project that could provide a window into the daily realities of life in different parts of the state, especially in Oregon’s hundreds of small towns.
In an increasingly urban state and nation, rural residents often find themselves unseen and unheard in the systems and decisions that affect their daily lives. Rather than focusing on what unites rural and urban experiences, mainstream attention centers on what divides us. We hope the Oregon Voices data can begin to shift that – dispelling divisive narratives and stereotypes within our state. We know that our fate is shared, rural or urban, and we need to see and appreciate each other’s lived experiences to set a course that serves all of us.
In most survey research conducted in Oregon, the methodology results in rural people and places being underrepresented. Because so much of Oregon’s population is concentrated in the Portland metropolitan area, studies using traditional representative sampling yield findings heavily weighted toward an urban point of view. To analyze the data by geography, many counties must be combined into regions – such as the Willamette Valley or Eastern Oregon. This practice means rural data get suppressed or excluded from statewide reports to protect confidentiality in communities where there are small populations.
By contrast, the first Oregon Voices survey used a methodology aimed at amplifying and elevating voices in the state’s less densely populated areas. We started by randomly selecting and inviting 500 households from every Oregon county to participate – whether the population of the county was 8,000 or 800,000. This ensured representation by county rather than by population distribution.
Two Phases Of Engagement
Offered in Spanish as well as English, online and on paper, there were two phases of data collection. For the first phase, the “Random Household” phase, 18,000 randomly selected households received multiple invitations to participate. We started with an email invite to all houses for which we had email addresses. We followed up with a physical letter from the Foundation delivered by mail, both encouraging completion of the survey online. After a few weeks, those who had not completed the survey online received a postal packet containing the print survey in English and Spanish with pre-paid return postage as well as information about a gift card drawing for anyone who submitted a survey. After a few more weeks, we sent out a final postcard reminder.
Then we launched what we call the “Ford Family” phase, where we extended the invitation to our grantees, scholarship recipients, community builders and regional partners to take part in the survey in print or online. We also encouraged them to share Oregon Voices with others they thought the Foundation should hear from.
Communications during both research phases included a phone number and email for contacting the Oregon Voices team for additional support. In the end, a total of 4,359 Oregonians completed the survey; 2,138 in the Random Household phase (roughly 50% completed via paper) and 2,221 in the Ford Family phase (all but a handful completed online).
Participation by County
Phase 1: Random Household
Phase 2: Engagement
Interpreting Survey Findings
There are a few nuances to keep in mind when exploring the Oregon Voices data. First, the overall data set is not representative of the population of Oregon. It is purposefully over-indexed for rural counties and communities. Second, when viewing the county-level data and summaries, it is important to note the proportion of respondents who participated from each phase. The people who participated during the Random Household phase tend to be older, have lower incomes, and have completed fewer years of schooling than those who participated during the Ford Family phase.
The interactive data tool allows users to explore these two groups and many others by filtering the data across eight different categories: county, geography (urban, suburban, rural), age, gender, ethnicity, income, education and respondent group.
This project would never have been possible without the generous responses of the thousands of Oregon residents who took part in our survey. We thank each and every one.
In addition, two organizations took the lead on entering, cleaning, coding and conducting early analyses of all the data that resulted – ECONorthwest and Portland State University. We are immensely grateful for their expertise and support. Our three organizations will continue to collaborate on the release of research briefs and trend reports in the months ahead, providing additional insights into the survey findings.