Survey Recruitment on Facebook and Instagram

Using Facebook and Instagram to Recruit Web Survey Participants: A Step-by-Step Guide and Application

1. Summary

Kühne and Zindel (2020) conducted a study on survey recruitment using social media platforms like Facebook and Instagram. An inevitable truth is, survey response rates are decreasing rapidly in many countries and contexts. The data collection is even more complex and expensive for hard-to-reach populations. Kühne and his team successfully utilized social media platforms for web survey recruitment on the LGBTQ populations in Germany, in which the respondents were truly willing to take the survey for free since they wanted their voice heard.

This study also provided a step-by-step guidance for future researchers on how to conduct a web survey using Facebook and Instagram. Since BEV owners are also sore of hard to reach, and these pioneers also want their voice heard, this study has truly provided an inspiration for our research, and has been providing contributive results.

2. Evidences

The Table 1 below shows past web surveys using Facebook and/or Instagram. This is a proof of successful application of social media recruitment.

Table 1: Web Surveys Using Facebook and/or Instagram — **Table 1**: *Web Surveys Using Facebook and/or Instagram*

The Table 2 below shows the overall cost and reach of the individual ad sets. On the controlling dashboard of Meta, each advertisement is deployed individually with reach number, click number, and cost clearly shown. It can be seen that the cost per click is relatively low. This table does not reveal the cost per complete response, but it’s not hard to calculate and is also inexpensive compared with panel recruitment.

Table 2: Overall Cost and Reach of Ad Sets — **Table 2**: *Overall Cost and Reach of Ad Sets*

Gaps and Improvements

The authors stated the following gaps:

Social media recruitment may not reach all target populations due to limited internet usage among some groups. Those who don’t use Facebook or Instagram are systematically filtered out.
The lack of transparency about Facebook's ad distribution algorithm is a concern.
Social media recruitment often relies on convenience samples, which can introduce bias due to unknown selection probabilities and self-selection of respondents.

The authors also stated the following future improvements to possible solve these gaps:

Comparing data from social media-recruited surveys with that from probability-based samples could help validate findings.
Integrating data from both probability and non-probability samples can potentially reduce bias and improve survey accuracy.

Probability-based vs Non-probability-based

Recruitment on Facebook and Instagram is considered as a combination of “probability-based” and “non-probability-based” sampling. It’s probability based due to the active selection of designated population, including location, age, gender, and specific groups that we are interested in. However, due to the fact that not everyone uses Facebook, and Facebook has its own algorithm of filtering, the recruitment is then distributed based on accessibility and convenience, which is why it’s considered “non-probability” as well.

References

Kühne, Simon, and Zaza Zindel. 2020. “Using Facebook and Instagram to Recruit Web Survey Participants: A Step-by-Step Guide and Application.” Survey Methods: Insights from the Field (SMIF), December. https://doi.org/10.13094/SMIF-2020-00017.

--- title: "Survey Recruitment on Facebook and Instagram" subtitle: "Using Facebook and Instagram to Recruit Web Survey Participants: A Step-by-Step Guide and Application" bibliography: ../references.bib --- ## 1. Summary @kühne2020 conducted a study on survey recruitment using social media platforms like Facebook and Instagram. An inevitable truth is, survey response rates are decreasing rapidly in many countries and contexts. The data collection is even more complex and expensive for hard-to-reach populations. Kühne and his team successfully utilized social media platforms for web survey recruitment on the LGBTQ populations in Germany, in which the respondents were truly willing to take the survey for free since they wanted their voice heard. This study also provided a step-by-step guidance for future researchers on how to conduct a web survey using Facebook and Instagram. Since BEV owners are also sore of hard to reach, and these pioneers also want their voice heard, this study has truly provided an inspiration for our research, and has been providing contributive results. ## 2. Evidences The **Table 1** below shows past web surveys using Facebook and/or Instagram. This is a proof of successful application of social media recruitment. ::: {style="text-align: center;"} ![**Table 1**: *Web Surveys Using Facebook and/or Instagram*](files/14_figure_01.png){width="800"} ::: The **Table 2** below shows the overall cost and reach of the individual ad sets. On the controlling dashboard of Meta, each advertisement is deployed individually with reach number, click number, and cost clearly shown. It can be seen that the cost per click is relatively low. This table does not reveal the cost per complete response, but it's not hard to calculate and is also inexpensive compared with panel recruitment. ::: {style="text-align: center;"} ![**Table 2**: *Overall Cost and Reach of Ad Sets*](files/14_figure_02.png){width="800"} ::: ## Gaps and Improvements The authors stated the following gaps: 1. Social media recruitment may not reach all target populations due to limited internet usage among some groups. Those who don't use Facebook or Instagram are systematically filtered out. 2. The lack of transparency about Facebook\'s ad distribution algorithm is a concern. 3. Social media recruitment often relies on convenience samples, which can introduce bias due to unknown selection probabilities and self-selection of respondents. The authors also stated the following future improvements to possible solve these gaps: 1. Comparing data from social media-recruited surveys with that from probability-based samples could help validate findings. 2. Integrating data from both probability and non-probability samples can potentially reduce bias and improve survey accuracy. ## Probability-based vs Non-probability-based Recruitment on Facebook and Instagram is considered as a combination of "probability-based" and "non-probability-based" sampling. It's probability based due to the active selection of designated population, including location, age, gender, and specific groups that we are interested in. However, due to the fact that not everyone uses Facebook, and Facebook has its own algorithm of filtering, the recruitment is then distributed based on accessibility and convenience, which is why it's considered "non-probability" as well.