Skip to main content
. Author manuscript; available in PMC: 2017 Jun 1.
Published in final edited form as: J Immigr Minor Health. 2016 Jun;18(3):561–567. doi: 10.1007/s10903-015-0240-y

Table 3.

Repeated Multivariate Analysis of VEE with Demographics, Alcohol Use and Psychosocial Variables in a Linear Mixed Model

Independent Variables VEE coefficient (SE)
Intercept 32.949 (4.049)***
Age −1.039 (0.838)
Ethnicity (African Americans) −4.440 (2.020)**
Education −1.020 (1.029)
Language (English) 2.929 (2.020)*
Social Provision Scale 1.929 (1.029)
Behavioral Intention −1.272 (1.029)***
Self-efficacy −2.920 (2.029)**
Condom Social Norm −0.842 (0.436)**
Short Inventory of Problems - R 1.983 (0.829)*
Alcohol use 0.672 (1.019)**
Intervention (P.E.E.R.) −1.219 (1.210)*
−2 Log Likelihood 6575.992 (df = 22)
Akaike’s Information Criterion 6589.099
Schwarz’s Bayesian Criterion 6599.020

Note. Higher the VEE, higher the sexual risk behavior. Social provision scale score ranged from 24 to 120; the higher score means higher level of social support. Behavioral intention score ranged from 1 to 4: the higher score means higher level of intentions to reduce HIV transmission risk behavior. Self-efficacy score ranged from 1 to 5; the higher score means greater confidence in negotiating condom use with partners in a variety of challenging sexual situations. Condom social norm score ranged from 1 to 4 and measured frequency at which respondents, as well as their family and friends, condoned condom use in a variety of ways. Higher score means higher condom social norm. Short Inventory of Problems - Revised score ranged from 17 to 68; the higher score means higher adverse consequences for alcohol and drug use. Alcohol use is the number of drinks in the last past month. The reference groups for the linear regression analysis are: ethnicity (Hispanics), intervention (H.E.A.L.T.H.Y.) and language (Spanish).

*

p < 0.05,

**

p < 0.01,

***

p < 0.001