Skip to main content
. 2022 Mar 15;24(3):e32678. doi: 10.2196/32678

Table 6.

Standardized coefficients and odds ratios for theorized OHISa models (for Latine and non-Latine individuals)b.

Path Group

Latine Non-Latine

Standardized coefficientc or odds ratiod (95% CI) P valuee Standardized coefficient or odds ratio (95% CI) P value
Predisposing characteristics → OHIS

Age (years) 0.979f (0.966-0.993) .002 0.977f (0.971-0.983) <.001

Sex 0.601f (0.474-0.762) .009 0.656f (0.581-0.741) <.001

Income 1.058f (0.930-1.204) .42 1.123f (1.051-1.200) <.001

Education 1.393f (1.075-1.804) <.001 1.481f (1.291-1.699) <.001
Predisposing characteristics → access

Home computer


Age (years) –0.185f <.001 –0.054g <.001


Sex 0.060f .08 0.063f <.001


Income 0.181f <.001 0.203f <.001


Education 0.324f <.001 0.254f <.001

Work computer


Age (years) –0.249f <.001 –0.374g <.001


Sex 0.019f .57 –0.006f .68


Income 0.265f <.001 0.301f <.001


Education 0.259f <.001 0.171f <.001

Public computer


Age (years) –0.257f <.001 –0.188f <.001


Sex –0.024f .57 0.033f .048


Income –0.152f .002 –0.142f <.001


Education 0.087f .10 0.128f <.001

Mobile


Age (years) –0.462f <.001 –0.439f <.001


Sex –0.069f .03 –0.072f <.001


Income 0.139f <.001 0.232f <.001


Education 0.206f <.001 0.114g <.001
Access → OHIS

Home computer 1.294f (0.861-1.945) .04 1.950g (1.541-2.467) <.001

Work computer 1.121f (0.772-1.627) .19 0.917f (0.806-1.044) .49

Public computer 1.500f (0.574-3.919) .08 1.117f (0.824-1.513) .15

Mobile 1.739f (1.057-2.861) <.001 1.912f (1.491-2.452) <.001
Predisposing characteristics → health need

Age (years) 0.140f <.001 0.093f <.001

Sex –0.039f .30 0.028f .07

Income –0.178f <.001 –0.243f <.001

Education –0.147f <.001 –0.113f <.001
Health need → OHIS 1.091f (0.869-1.369) .58 1.242f (1.093-1.411) .001

aOHIS: online health information seeking.

bComparisons were made for each model per row.

cStandardized coefficients are displayed for paths predicting nondichotomous outcomes; negative relationships are indicated by negative signs. For standardized coefficients, 95% CI values are not available.

dOdds ratios are presented for paths predicting dichotomous outcomes (ie, OHIS) and were generated using Monte Carlo integration because of model complexity; negative relationships are indicated by values <1.

eSignificance values were based on the primary models (ie, without Monte Carlo integration).

fCoefficients or odds ratios differ significantly from those denoted by footnote g at P<.05.

gCoefficients or odds ratios differ significantly from those denoted by footnote f at P<.05.