Table 3.

Study 1: Regression results for perceived consequences of fast food

	Unstandardized Regression Coefficients (Standard Errors)

	Positive perceptions	Negative perceptions

(Constant)	1.920 (.445)^**	4.09 (.501)^**
Fast-food experience	.118 (.044)^**	.052 (.050)
TV viewing^a	.010 (.023)^**	−.081 (.026)^**
Fast-food experience × TV viewing	−.004 (.004)	.037 (.020)^b
Physical Activity	−.010 (.006)	.005 (.007)
Gender (1 = Male / 0)	.190 (.096)^*	−.247 (.108)^*
Year of birth	−.008 (.055)	−.032 (.062)
Ethnicity (1 = White / 0)	.408 (.274)	−.090 (.309)
Ethnicity (1 = Black / 0)	.734 (.310)^*	−.482 (.350)
Ethnicity (1 = Hispanic / 0)	.489 (.341)	−.652 (.385)
Ethnicity (1 = Asian American / 0)	.957 (.393)^*	.302 (.443)
Subjective SES	.058 (.026)^*	−.052 (.029)

Adjusted R²	.101	.045

^**

p ≤ .01,

p ≤ .05.

TV viewing, which had a larger variance than the other variables, was divided by 10 before being mean-centered. Results did not change if analyses were conducted with the weekly viewing amount of TV series (sitcoms, soap operas, dramas, reality TV) instead of overall TV viewing or with an alternate self-report measure computed as the sum of weekday and weekend viewing. Because cultivation research uses overall TV viewing, this measure was preferred and therefore is reported in the table.

p = .06, but all relevant contrasts are significant.