Why Just Survey US-Based Writers!!!

Free AI web copilot to create summaries, insights and extended knowledge, download it at here

1396

Abstract

have been done.</p><ul><li>They could only afford to do the survey in one country and the US is their biggest market</li><li>They plan to do the survey in other countries but need to spread the cost over more than one fiscal quarter — or the group with the purse strings wants to see how things pan out in the first country before authorizing more money for more countries (this is typical for companies who are rather new to marketing research)</li></ul><h2 id="31ce">Why an Amazon Gift Code?</h2><p id="5468">Others expressed concern over getting a gift code for the $5 incentive. Why not an account credit? Again, there is a couple of reason for this:</p><ul><li>Since a third-party research partner is fielding the research, there would need to be communication between them and Medium as to who completed the survey. This runs counter to the privacy code for the Marketing Research industry.</li><li>Allowing the research company access to the accounts of Medium writers/subscribers likely goes against Medium’s privacy policy.</li><li>Amazon codes are much easier and cheaper to administer than mailing out incentive checks or physical cards.</li></ul><h2 id="0159">Why Have Quotas for Various Demographics?</h2><p id="d805">First of all, you go into a survey with a targeted number of completes in mind. As a third-party research partner (when I was still in business), the ending sample s

Options

ize is one driver of the project cost. How one reaches the target number can vary according to budget. In terms of research design and analysis, more is better, but with diminishing returns. The total sample size should be at least 100 people. Better is a sample of 300; better still is a sample of 500. Ten thousand is too many… One driving factor for sample size is whether the results will be analyzed in total, or if subgroups will be analyzed (e.g., male Vs. female writers). If subgroups are to be analyzed separately, you start with your smallest group at n = 100 and work up, proportionately.</p><p id="d812">Either way, it’s important that the ending sample of respondents to your survey reflect the various demographic groups present in the platform “population.” But you want the reflection each group makes to be proportional to their presence in the population. So, if 70% of US-based writers are female, you would want to limit participation by male writers to 30% of the targeted number of completed surveys. The same goes for age groups.</p><h2 id="f06b">Summary</h2><p id="62ce">In the end, there are numerous reasons for doing a survey among only US-based writers — none of which have anything to do with conspiracy theories or discrimination or ageism or sexism or xenophobia or being stupid. That so many jumped to such conclusions was, for me, surprising.</p></article></body>

Why Just Survey US-Based Writers!!!

No, I don’t work for Medium or their Marketing Research partner

It’s probably not the USA is the #1 thing (Licensed via Freepik.com)

A lot of dust has been brushed up over the recent survey Medium is conducting among US-based writers on the platform. Some seem to think the focus is US-based because Medium is run by a bunch of kids in high school who are, apparently xenophobic, racist, ageist, sexist people. And this could be true. OR, maybe they just hired a Marketing Research firm that follows best practices in conducting surveys.

As background, I have been conducting marketing research for companies around the world for over 30 years — in fact, I just retired a couple of months ago. My company website is still up (although the business has been closed down) — see www.ergointernationalinc.com if you are interested. I’ve done several multi-country studies for the likes of Apple, Spotify and Samsung.

Why US-Based Only?

Some of the dust was raised over including only US-based writers in the survey. There are a couple of reasons this could have been done.

They could only afford to do the survey in one country and the US is their biggest market

They plan to do the survey in other countries but need to spread the cost over more than one fiscal quarter — or the group with the purse strings wants to see how things pan out in the first country before authorizing more money for more countries (this is typical for companies who are rather new to marketing research)

Why an Amazon Gift Code?

Others expressed concern over getting a gift code for the $5 incentive. Why not an account credit? Again, there is a couple of reason for this:

Since a third-party research partner is fielding the research, there would need to be communication between them and Medium as to who completed the survey. This runs counter to the privacy code for the Marketing Research industry.

Allowing the research company access to the accounts of Medium writers/subscribers likely goes against Medium’s privacy policy.

Amazon codes are much easier and cheaper to administer than mailing out incentive checks or physical cards.

Why Have Quotas for Various Demographics?

First of all, you go into a survey with a targeted number of completes in mind. As a third-party research partner (when I was still in business), the ending sample size is one driver of the project cost. How one reaches the target number can vary according to budget. In terms of research design and analysis, more is better, but with diminishing returns. The total sample size should be at least 100 people. Better is a sample of 300; better still is a sample of 500. Ten thousand is too many… One driving factor for sample size is whether the results will be analyzed in total, or if subgroups will be analyzed (e.g., male Vs. female writers). If subgroups are to be analyzed separately, you start with your smallest group at n = 100 and work up, proportionately.

Either way, it’s important that the ending sample of respondents to your survey reflect the various demographic groups present in the platform “population.” But you want the reflection each group makes to be proportional to their presence in the population. So, if 70% of US-based writers are female, you would want to limit participation by male writers to 30% of the targeted number of completed surveys. The same goes for age groups.