This presents a perceptual map of voter attitudes on the seven ballot measures on the Oregon ballot.
There are several obvious clusters.
Measures 88 (driver card), 91 (legalize marijuana), and 92 (GMO labeling) are closely packed. This makes sense because they are the three most obviously liberal measures. The closest one to these is Measure 86 (college fund) which is also liberal leaning and then comes Measure 89 (Equal Rights Amendment).
The proximity of the two comparatively conservative measures, 87 (judicial employment) and 90 (top two primary) is better explained as due to their good government attributes as opposed to their conservative attributes.
Component 1 seems to distinguish the simple yes versus no decision. Component 2 seems best to distinguish between good government on the one hand and liberal on the other hand.
QUESTIONS INCLUDED: Now, I would like to talk to you about some measures that will be on the November ballot. If the election were held today on ________________, would you vote yes or no?
IF YES/NO: Is that strongly or somewhat?
IF DON’T KNOW: Which way do you lean?
1. Measure 86: Amends Constitution: Requires creation of fund for Oregonians pursuing post-secondary education, authorizes state indebtedness to finance fund
2. Measure 87: Amends Constitution: Permits employment of state judges by National Guard (military service) and state public universities (teaching)
3. Measure 88: Provides Oregon resident “driver card” without requiring proof of legal presence in the United States
4. Measure 89: Amends Constitution: State/political subdivision shall not deny or abridge equality of rights on account of sex
5. Measure 90: Changes general election nomination processes: provides for single primary ballot listing candidates; top two advance
6. Measure 91: Allows possession, manufacture, sale of marijuana by/to adults, subject to state licensing, regulation, taxation
7. Measure 92: Requires food manufacturers, retailers to label “genetically engineered” foods as such; state, citizens may enforce
METHODOLOGY: 400 live telephone interviews were conducted of likely State of Oregon voters October 13-16, 2014. Both landlines and cell phones were included. Likelihood was determined by a model overlaid on the voter file. The margin of error at the sample median is plus or minus five percent.
The perceptual map was constructed using factor analysis. Technically, it a Principal Components Analysis was used with unrotated factors. A seven-point scale for each of the questions was used. Factors were deemed significant if their eigenvalue was greater than one.
This methodology was used because it involved the least number of assumptions and, in that sense, was the “purest” way to obtain the results. There are more sophisticated methods that would arguably be more illuminating. The objective here was lack of bias and transparency.