1 of 18

Lenfest Midterm Elections Pre-Survey

October 1, 2018

Hero Digital

2 of 18

Analysis Overview

3 of 18

Summary

3

Majority of respondents were from PA and read PMN content

75% of respondents have been voting for 20+ years, results may be skewed towards a very politically active audience

Local news was the top source for voting related information

Opportunity exists with weekly issue-based news, however, respondents indicated concerns could include political bias and texts causing annoyance�
14% of respondents did not vote before because they felt that did not have enough information

Segmentation shows slightly different preferences so direction moving forward should be dependent on experiment goals

4 of 18

Respondent overview: Pennsylvania PMN Readers

4

254 total responses

91% of respondents lived in PA

90% of respondents read PMN content, possibly impacted by the way the survey was distributed
60% of respondents subscribe to PMN content
27% of respondents subscribe to both print and online content

5 of 18

Political Profile: Long-term politically active

5

Only 1 respondent hadn’t voted before

95% of respondents usually vote in all 3 main election types - national, state and local elections

75% of respondents have voted for 20+ years

6 of 18

Local news was top source for voting information

6

Majority of respondents get voting information from local, online research and national news

80% of respondents chose local news as a source of information

Only 4 respondents have never sought out information before

7 of 18

Opportunity with weekly issue based news

7

Although nothing has kept 70% of respondents from voting, respondents expressed most interest in issue-based news

83% of respondents stated that issue-based news would be most useful, followed by ballot questions

77% of respondents stated a preference for receiving the texts “weekly” or “a few times a week”

Some respondents mentioned that the cadence could increase closer towards the elections

8 of 18

Cadence will be crucial

8

50% of respondents said that they would be concerned that the texts would get annoying

A few of the long form answers mentioned that email would be a more suitable format

Easily allowing users to stop the texts or modify their preferences will prevent users from thinking the experiment is annoying

Bias was the most popular write-in other field (12 respondents)

Important to consider that issue-based information may appear to be more opinionated than topics such as ballot explainers or guide to election language

9 of 18

14% of respondents didn’t feel informed enough

9

Opportunity with this segment, more information could change their behavior

Despite previously not feeling informed enough, 75% of this segment has been voting for 10+ years

Most interested in issue based and ballot question explainers (30 respondents each)

Consistent with the overall group, 89% of this segment prefer to receive such information weekly or a few times a week

10 of 18

Segments Based on Voting History

11 of 18

New voters more likely to rely on TV as source of info

11

Local news, national news and online research are top sources across all voting length segments
Longer term voters stated gaining information from political events likely due to being more politically active
Note: segments may be misrepresentative as 75% of the respondents have been voting for 20+ years

12 of 18

Segments have slightly different topic preferences

12

Question: do we want to prioritize the largest type of respondents (20+ years) or focus on a smaller group such as 0 – 3 since there may be more opportunity to encourage voting more?
All segments mentioned issued based info as one of the most useful topics
New voters find guide to election language more useful than long-term voters
Interestingly, long term voters find ballot explainers to still be highly useful

13 of 18

All segments prefer a less frequent cadence

13

Majority of respondents from each segment had concerns that the texts would get annoying

Interestingly, only the new voter segment mentioned a concern of receiving too much information

14 of 18

Opinions from long form responses

14

Positive:

Sounds like a good service

It's definitely difficult to find information about local and state races, specifically about candidates…

We desperately need this. Few ppl read much these days and have very short attention spans. Texts are a fantastic way to reach people.

Great idea for people on the fence about voting

I think this could be very useful, and I'd like to be a beta tester if you develop a text-based election notification system.

Negative:

You are trying to influence an election and I don’t find this good. What will you do? Tell people who to vote for if they text back questions? I find this very offensive and dangerous…

15 of 18

Recommendations

16 of 18

Recommendations/Considerations

16

May be useful to distribute the survey to more new voters to ensure representative feedback

Need to determine if the focus is to create an experiment that best serves the most available audience (20+ years of voting) or to reach new voters who may need the information the most

Issue based information was the most popular type of content desired but it will be important to present objective/unbiased opinions

Aside from issue based information, other topics are more useful to some segments over others which is why the team will need to solidify the experiment goals. May be useful to allow users to customize amount of information desired

Since the majority of respondents stated concerns around annoyance, recommend delivering the texts at maximum once a week or would need a method to allow the user to customize the timing

17 of 18

Additional product ideas from long-form responses

17

More available here

For incumbents, voting record on major issues

Progress (or impediments) to getting early in-person voting in PA

Reach out to neighborhood stakeholders

Assessment of candidate claims and ads--fact checking

Reminder before the election on what day the election is, when the polls open, and sent a warning before they close

Side by side issue comparison of candidates’ platforms or policy positions

Information on voting for judges, current lack of information

18 of 18

Thank You