Advertising Surveys on Social Media

Advertising Online Surveys on Social Media: How Your Advertisements Affect Your Samples

1. Summary

Neundorf and Öztürk (2022) performed a study in 2022 investigating the impact of different advertising content on the recruitment of online survey participants via social media with focus on political science research. 22 advertising campaigns were conducted in Turkey and Spain, engaging over 30,000 respondents to compare different advertisement strategies. There are 3 main strategies: Neutral, Thematic, and Incentive-based. The incentive-based strategy produces the best performance, but the type of incentive should be carefully considered.

2. Advertisement Strategies

There are 3 strategies:

Neutral advertisements do not mention the political content of surveys and do not offer material incentives to social media users.
Thematic advertisements aim to draw the attention of social media users by explicitly mentioning the theme of the survey.
Advertisements with incentives offer material benefits to social media users in return for completing the survey.

3. Research Questions

Based on the 3 strategies, this paper asks 4 questions:

Which advertisement strategy does produce the cheapest sample?
Which advertisement strategy does produce the most representative sample?
Does advertisement content have any effect on response quality?
Which versions of these advertisement strategies do perform the best?

4. Results

Incentive-based ads can produce highly-representative samples, but effectiveness varies depending on the type of incentive offered. An incentive with large value but very low chance of winning (an iPad, for example) is surprisingly not as efficient as one with lower value but higher chance. Like the authors mentioned, the web users were already saturated with lotteries promising an iPad. A small fortune like a store gift card is rather more realistic.
Thematic ads (ads that mention the political content of the survey) are cheapest, but result in samples that over-represent certain demographics (men, older individuals, college graduates, politically interested individuals, and strong partisans).
Sadly, although the ad content is important to recruit participants, it does not have significant effect on the response quality.

To conclude, incentive-based ad is the best possible solution, but its effectiveness highly depends on the type of incentive offered. Thematic ads produce biased sampling, and neutral is not good in general.

References

Neundorf, Anja, and Aykut Öztürk. 2022. Advertising Online Surveys on Social Media: How Your Advertisements Affect Your Samples. https://doi.org/10.31219/osf.io/84h3t.

--- title: "Advertising Surveys on Social Media" subtitle: "Advertising Online Surveys on Social Media: How Your Advertisements Affect Your Samples" bibliography: ../references.bib --- ## 1. Summary @neundorf2022 performed a study in 2022 investigating the impact of different advertising content on the recruitment of online survey participants via social media with focus on political science research. 22 advertising campaigns were conducted in Turkey and Spain, engaging over 30,000 respondents to compare different advertisement strategies. There are 3 main strategies: Neutral, Thematic, and Incentive-based. The incentive-based strategy produces the best performance, but the type of incentive should be carefully considered. ## 2. Advertisement Strategies There are 3 strategies: 1. **Neutral** advertisements do not mention the political content of surveys and do not offer material incentives to social media users. 2. **Thematic** advertisements aim to draw the attention of social media users by explicitly mentioning the theme of the survey. 3. Advertisements with **incentives** offer material benefits to social media users in return for completing the survey. ## 3. Research Questions Based on the 3 strategies, this paper asks 4 questions: 1. Which advertisement strategy does produce the **cheapest** sample? 2. Which advertisement strategy does produce the most **representative** sample? 3. Does advertisement **content** have any effect on response **quality**? 4. Which versions of these advertisement strategies do perform the best? ## 4. Results 1. **Incentive**-based ads can produce highly-**representative** samples, but effectiveness varies depending on the type of incentive offered. An incentive with large value but very low chance of winning (an iPad, for example) is surprisingly not as efficient as one with lower value but higher chance. Like the authors mentioned, the web users were already saturated with lotteries promising an iPad. A small fortune like a store gift card is rather more realistic. 2. **Thematic** ads (ads that mention the political content of the survey) are **cheapest**, but result in samples that over-represent certain demographics (men, older individuals, college graduates, politically interested individuals, and strong partisans). 3. Sadly, although the ad **content** is important to recruit participants, it does **not** have significant effect on the response **quality**. To conclude, **incentive**-based ad is the best possible solution, but its effectiveness highly depends on the type of incentive offered. Thematic ads produce biased sampling, and neutral is not good in general.