Assessing effects of a media campaign on HIV/AIDS awareness and prevention in Nigeria: results from the VISION Project

Table 3 Poisson regression results (beta coefficients) estimating exposure to the number of media programs over the past 6 months (3 models)

Covariates (PSU = 120)	Total Programs (n = 3,249)	Radio (n = 3,249)	TV (n = 3,252)
Watch television once a week	0.237**	0.082	0.664**
Listen to the radio once a week	0.634**	0.787**	0.407*
Read newspaper once a week	0.174**	0.151*	0.296*
Age (continuous variable)	-0.002	-0.007**	0.015**
At least 1 partner in past 12 months	0.039	0.033	0.013
Education (No formal education: Reference)
Primary	0.139*	0.112	0.104
Secondary	0.270**	0.151*	0.257*
Residence (Rural: Reference)
Urban/peri-urban	0.193**	0.185	0.869**
Gender (Female: Reference)
Male	0.135**	0.108*	0.185*
Married or living with partner	0.013	0.104*	-0.289**
State (Oyo: Reference)
Bauchi	0.531**	0.578**	0.808**
Enugu	-0.103	-0.264*	-0.156
Constant	-0.643**	-0.958**	-3.493**

^§Weighted data were used for the Poisson regression; ^†Standard errors were estimated using a sandwich estimator; *Significant with p-value <0.05; **Significant with p-value <0.001; Note: The values indicated in the table represent the estimated beta coefficients. A positive beta coefficient indicates a positive association with number of program exposures. The constant term represents the direction and magnitude of the relationship for the respective reference categories.

ISSN: 1471-2458