Supporters of Initiative 82 (the DC ballot measure that would end the tipped minimum wage/tip credit for good by July 2027 if passed) point to a survey of 300 DC tipped workers that showed 91% support ending the tipped minimum wage. Those opposed appear to dismiss the results out of hand, and say they haven’t talked to any tipped workers who support it.

I haven’t seen really anyone try to dig into these survey results, and thought perhaps it would be at least somewhat illuminating to break it down and also provide some details about how the survey was conducted, based on a few emails I exchanged with with one of the survey administrators.

Who Conducted the Survey?

The survey was conducted by Lake Research Partners (LRP), presumably commissioned by One Fair Wage (I presume this because the presentation publicizing the data is posted on their website and includes their logo), which is the advocacy organization trying to end tipped minimum wage across the United States. LRP is, according to its website: “a national public opinion and political strategy research firm founded by Celinda Lake in 1995, and the most consistently accurate—and consistently progressive—Democratic research firm in the country.” Five-Thirty-Eight gave LRP a A/B provisional rating for their political polls.

When Was the Survey Conducted?

July 1-7, 2021, with results released in September 2021.

How Was the Survey Conducted?

The survey was administered and conducted online with 300 servers/bartenders who work in DC, according to the presentation. This seemed pretty vague, so I reached out directly to LRP for additional information:

LRP (like other survey pollsters) reach out to panel providers/vendors who provide panels of individuals to complete their surveys. These companies generally (by lots of different online recruitment means) collectively recruit tens of thousands of people across the country who volunteer to be sent surveys.

For a survey like this, a LRP staffer explained to me “For this survey, respondents would have been pre-flagged to be tipped workers by the panel providers during the universe design phase where we are developing a subset of the panels to send invitations to. Then they will also have to screen in by self-identifying as someone who has worked as a bartender, waiter, or server within the last 18 months (for more than 10 hours a week).” Everyone pre-flagged would receive a “unique link” that cannot be forwarded to anyone else.

The staffer explained to me that the survey was distributed to a “subset of panel participants, which is stratified by demographic variables (including age, gender, race, and education level) to match the expected proportions in the universe (in this case DC restaurant workers).“

They also explained that “We also set quotas on demographics to ensure that from the pre-stratified sample certain groups who are more likely to respond to surveys will not be overrepresented, and then when interviewing is complete we use weighting as necessary, to ensure that the final demographic distribution of our survey sample matches our expected demographic profile of DC restaurant workers.”

So, What Was the Breakdown of Tipped Workers Who Responded?

According to LPR, those who participated in the survey self-identified as the following:

34% bartenders

51% server/waiter

15% both

What Questions Did the Survey Ask?

Based on what was released :

• 1) As you may know, there is currently a proposal in the District of Columbia to increase the minimum wage over several years for tipped workers to $15 an hour with tips on top. Most restaurant workers currently are paid lower than the full minimum wage, with the difference made up by tips or by their employer. This proposal would raise the minimum wage for tipped workers to the full minimum wage while they retain tips on top. [Then likely asked: Do you favor or oppose the proposal to raise the minimum wage for tipped workers to the full minimum wage while they retain tips on top?”]
• 2) Now you’re going to get a little more information about this proposal: There are already seven states that require tipped workers receive the full minimum wage with tips on top. Data from credit card processing shows that restaurant workers in these seven states earn the same or higher tips as workers in other states. Sometimes in a survey like this, people change their minds. Do you favor or oppose the proposal to raise the minimum wage for tipped workers to the full minimum wage while they retain tips on top? [This question was “split sampled” meaning that not all 300 workers got this question worded this way]
• 3) If a candidate for office was proposing to raise the minimum wage for tipped workers to the full minimum wage with tips on top, would that make you more or less likely to vote for that candidate, or does it make no difference to you?

What Were the Published Results?

• Question 1: 54% Strongly Favor, 34% Strongly Favor. 12% Oppose or Strongly Oppose
• Question 2: 50% Strongly Favor, 41% Favor. 9% Oppose or Strongly Oppose
• Question 3: 43% Much More Likely, 36% More Likely, 17% Less Likely or Much Less Likely, 3% No Difference
• “Base sample” margin of error was +/- 5.7%

Anything Else?

At least a couple hospitality workers opposed to Initiative 82 have complained/speculated that One Fair Wage/pro I 82 folks stacked the deck somehow and somehow got only pro I 82 folks to complete the survey. There’s no evidence of that based on how LRP explained.

Some of that speculation might come from the fact that One Fair Wage established an Emergency Fund to help workers affected by the pandemic; over 260,000 applied. All those applicants were emailed surveys by One Fair Wage/UC Berkeley Food Labor Research Center starting in October 2020 (over 4,300 responded) regarding about restaurant worker retention. But this LPR survey does not appear to be connected to that one. I suppose One Fair Wage could have theoretically emailed all those people and told them to sign up for panel providers in the chance they could at some point be asked to participate in a survey about the tipped minimum wage, but that seems pretty far-fetched.

LRP told me there were way more than 300 DC tipped workers in the panel providers they reached out to who opted in to participate in polls.