Sources
Verification note. Full topline (verbatim wording, frequencies and percentages) and the methodology paragraph read directly off Houston Public Media's own poll page. FIELD DATES ARE NOT PUBLISHED: the pollster gives no interviewing dates. The window is bounded by the design - respondents had to have voted on November 3, 2015 and some had already voted early in the December 12 runoff, early voting having opened November 30 - and by the December 10, 2015 release; field_start/field_end are therefore approximate and date_precision is 'month'. This supersedes the earlier unverified DB records 'unknown_2015-11_houston-texas' and 'bob-stein_2015-12_houston-texas'. NO CROSSTABS: the page publishes marginal frequencies only, so no party breaks on the HERO items are available; party identification in the sample as a whole was 36% Democrat, 29% Republican, 26% Independent, 9% no answer. BALLOT BENCHMARK, VOTE-RECALL ITEM: the second question below is a recall of the actual November 3, 2015 vote. Excluding don't-know and refused, recalled vote was 34.8% For / 65.2% Against, against a real result of roughly 39% For / 61% Against - i.e. the sample over-reports having voted to repeal by about four points, the usual bandwagon direction. BALLOT BENCHMARK: Houston Proposition 1 (the HERO referendum) was voted on November 3, 2015. Voters repealed the ordinance; the measure failed roughly 61 percent 'Against' to 39 percent 'For'. Exact certified vote totals were not retrieved in this session. || Full topline (verbatim wording, frequencies and percentages) and the methodology paragraph read directly off Houston Public Media's own poll page. FIELD DATES ARE NOT PUBLISHED: the pollster gives no interviewing dates. The window is bounded by the design - respondents had to have voted on November 3, 2015 and some had already voted early in the December 12 runoff, early voting having opened November 30 - and by the December 10, 2015 release; field_start/field_end are therefore approximate and date_precision is 'month'. This supersedes the earlier unverified DB records 'unknown_2015-11_houston-texas' and 'bob-stein_2015-12_houston-texas'. NO CROSSTABS: the page publishes marginal frequencies only, so no party breaks on the HERO items are available; party identification in the sample as a whole was 36% Democrat, 29% Republican, 26% Independent, 9% no answer. BALLOT BENCHMARK, VOTE-RECALL ITEM: the second question below is a recall of the actual November 3, 2015 vote. Excluding don't-know and refused, recalled vote was 34.8% For / 65.2% Against, against a real result of roughly 39% For / 61% Against - i.e. the sample over-reports having voted to repeal by about four points, the usual bandwagon direction. BALLOT BENCHMARK: Houston Proposition 1 - the referendum on the Houston Equal Rights Ordinance - was voted on November 3, 2015. Certified result (Harris County Elections Office cumulative report, November 11, 2015): Yes (keep HERO) 100,582 = 39.03%; No (repeal) 157,110 = 60.97%. Valid votes 257,692; turnout 27.45% of 979,401 registered voters. POLLING ERROR ON THE RECALL ITEM: excluding don't-know and refused, recalled vote was 145 For / 272 Against = 34.77% / 65.23%, against a certified 39.03% / 60.97%. The sample over-reports having voted to repeal by about 4.3 points on each side - the ordinary bandwagon direction, and a useful size estimate for recall bias on a contested trans-rights ballot measure.
Figures below are transcribed from these documents. Where they disagree with the source, the source governs.