When is a reminder enough? Testing the effectiveness of text messaging for mobilizing voters in a European context

One voter mobilization technique that has been tested to only a limited degree is sending text messages reminding voters of an upcoming election. The existing research has led to the formulation of the Noticeable Reminder Theory (Dale & Strauss 2009; Malhotra 2011). The theory’s premise is that registered voters generally have the intention to vote, but frequently fail to do so because of time constraints and lack of planning. To mobilize these voters normative arguments in favor of voting are unnecessary as voters readily acknowledge their intention to vote; all voters need is a noticeable reminder, such as one provided by a text message just prior to an election. Our empirical test provides a strong test of theory robustness. Previous tests have been done almost exclusively in the United States. We test under a dramatically different set of conditions in the Norwegian context. We received voter registration records across 27 municipalities for just over 1.7 million voters (42% of the voters the country). For approximately 80 percent we were also able to get mobile telephone numbers. This group was stratified both with respect to geography and with respect to voter background (first time eligible foreign nationals, foreign nationals who have been eligible for at least six years to vote, and Norwegian natives). We sent out about 130,000 text messages (max 140 characters) reminding people to vote. Half of the messages were split evenly across each day in the last week before the election, while the rest were sent out at hourly intervals on election day. The results of the experiment were strikingly strong. SMS text messages have an "Intention to treat" (ITT) effect of .96 percent among “attentive voters”, who we define as voters over the age of 30 (baseline turnout was .72 in this group). The messages are more effective in the group of foreign nationals who have become newly eligible to vote in local elections (having met the three years of legal residence requirement sometime after the last local elections). We expected this group to be largely inattentive as many are likely to have limited knowledge of the Norwegian language, and thus pay little attention to the country’s media. Nevertheless, the text messages produce an ITT of 3.13 percentage points (with a baseline turnout of 22 percent). The strongest effects, however, are among the two groups of sporadically attentive voters: an ITT of 4.58 percentage points among Norwegians below the age of 30, and 3.32 among people with immigrant backgrounds but who have been eligible to vote for greater than seven years in Norway. Text message effectiveness did not vary by time. Messages sent a full week before the election were just as effective as messages sent the day of the election. These results indicate that SMS text messages can be very effective tools for mobilizing voters including those who tend to participate less than the typical voter. The implications of our findings for mobilization theory are extensively discussed.