The authors examined the reliability and validity of the emotional and instrumental social support scale and Berkman's Social Network Index(SNI), both of which had developed for the western, middle class to test their adequacy for the Korean in
the
baseline data of the on-going breast-feeding discontinuation follow-up study. The subjects were recruited 525 wmen who delivered their babies in the Dongsan university Medical Center and one OB/GYN hospital located in Taegu city September to
November
1991. For mothers who agreed to enter the follow-up study administered were the questionnaires containing ssociodemographic variables and social support scale adopted from Strogatz and James and SNI from Berkman's before discharge. Mean age and
educational attainment were 27.3(standard deviation, SD 2.2) and 13 years(SD, 2.4) respectively. Most of the subjects(468, 89.1%) were reported to have adequate instrumental support and 348(66.7%) to have adequate emotional support. Correlation
coefficients among social support items and SNI items were 0.36-0.62 and 0.04-0.31, respectively. Those between items of social support and SNI were 0.00-0.15. Internal consistency was 0.75 for social support scale and 0.36 for SNI. And religious
attendance of SNI was poorly predicted from the other items. Factor analysis with varimax rotation yielded factor I consisted of social support scale items and factor II of SNI items. Communality of items were 0.502-0.672 except 0.116 for marital
status
and 0.097 for religious attendance. Factor loadings of marital status and religious attendance were 0.335 and 0.311 respectively, which were much lower than other items(0.691-0.818). Multiple logistic regression was performed to examine the
sociodemographic variables associated with instrumental and emotional social support(low=0, adequate=1). Yonger age(B=-0.796, standard error(SF)=0.227, p=0.00) and protestant in contrast with no religion (B=1.150, SE=0.573, p=0.04) were
associated
with
adequate instrumental support after controlling for educational attainments, marital status, types of social security, place of residency and status of employment. Those who reported adequate emotional social support were more educated(B=0.254,
SE=0.108, p=0.02) and unemployed(B=-0.623, SE=0.282, p=0.03). Multiple regression performed to examine the sociodemographic variables associated with SNI showed that more educated mothers reported to have wide SNI(B=0.164, SE=0.072, p=0.02).
Instrumental and emotional social support scales seemed to be adequate for our subjects but SNI appeared to be inadequate, particularly religious attendance item. Which might reflect culture-specific aspects of SNI developed for the western,
white
middle class different form the Korean's
|