Cross-tabs! We have cross-tabs!

Here are the breakouts for the WCPO/Survey USA poll of 504 likely voters conducted Aug. 27 to Aug. 29 on the Cincinnati mayoral primary:

	Mallory	Pepper	Reece	Winburn	Other	Unsure
All likely voters
Total	21%	29%	28%	15%	3%	3%
By sex
Male	23%	28%	26%	16%	4%	3%
Female	19%	29%	29%	15%	3%	4%
By age
18-34	23%	20%	40%	9%	3%	3%
35-49	21%	22%	30%	17%	4%	6%
50-64	19%	38%	22%	15%	3%	3%
65+	21%	40%	15%	21%	3%	1%
By race
White	18%	48%	11%	15%	4%	4%
Black	26%	7%	48%	14%	2%	3%
Hispanic	17%	45%	13%	8%	17%	0%
Other	11%	20%	27%	29%	9%	4%
By party affiliation
Republican	11%	48%	8%	27%	5%	1%
Democratic	26%	21%	38%	9%	2%	4%
Independent	21%	28%	26%	17%	6%	3%
Not Sure	20%	18%	5%	36%	0%	21%
By education
Grad School	27%	40%	13%	11%	4%	5%
College	25%	31%	19%	17%	5%	4%
Some College	21%	24%	33%	15%	4%	4%
No College	16%	27%	35%	17%	2%	2%
By ideology
Conservative	11%	31%	22%	29%	5%	2%
Moderate	22%	32%	27%	14%	2%	4%
Liberal	31%	25%	32%	6%	4%	2%
Not Sure	16%	13%	49%	4%	3%	14%

The big story here is that Democrat David Pepper is doing even better among Republicans than the endorsed Republican candidate, Charlie Winburn, 48 percent to 27 percent. That's a big jump from the Aug. 1 WCPO poll, when Pepper and Winburn split the GOP vote more or less evenly.

Even among self-described "conservatives," Pepper and Winburn are in a statistical tie. That's why Winburn keeps touting his Fraternal Order of Police endorsement and identifying himself as the only endorsed Republican in the race.

But Pepper continues to struggle among black voters, getting 7 percent of the African-American vote in a field with three strong black candidates as alternatives.

Alicia Reece and Pepper are tied for the lead, but it's worth noting that Reece's support comes from demographic groups -- young people, the lesser educated and those "not sure" of their ideology -- who are least likely to vote, especially in a mayoral primary with nothing else on the ballot. Expect the Reece camp to concentrate on get-out-the-vote efforts in the next two weeks.

The usual methodological caveats apply: In addition to the statistical margin of error (plus or minus 4.5 percentage points for the overall results, higher for demographic sub-groups) the poll could have biases inherent in touch-tone polls (e.g., a 5-year-old who can answer the phone can pass herself off as a "likely voter.") Also, the number of undecided voters -- 3 percent -- defies credibility.

Still, WCPO is the only news organization doing scientific polling in the mayor's race, so it's the best data we have.