Duke-UNC Functional Social Support Questionnaire Broadhead, W.E., Gehlbach, S.H., DeGruy, F.V., and Kaplan, B.H. 1988 Description of Measure Purpose To measure an individual s perception of the amount and type of personal social support. Conceptual Organization The original instrument included 14 items, grouped into 4 subscales: Quantity of Support, Confidant Support, Affective Support, and Instrumental Support. Item Origin/Selection Process The 14 items were derived from a larger questionnaire developed from a review of the literature for content validity and pretested for reliability (Broadhead, Gehlbach, DeGruy, & Kaplan, 1988, 1989). Materials See Broadhead, Gehlbach, DeGruy, & Kaplan, 1988. Time Required 5 minutes Administration Method Interviewer- or self-administered Training Minimal Scoring Score Types The item response options are on a 5-point scale ranging from 1 (much less than I would 201
like) to 5 (as much as I would like). Score Interpretation Higher scores reflect higher perceived social support. Norms and/or Comparative Data The Duke-UNC Functional Social Support Questionnaire (FSSQ) was developed and tested on 401 randomly selected patients attending a family medical clinic in Durham, NC. The population was primarily white, female, and of high SES. Mean item scores on the 14-item instrument for this group ranged from 3.54 to 4.34 on a 5-point scale (Broadhead, Gehlbach, DeGruy, & Kaplan, 1988). Psychometric Support Reliability Test-retest reliability was evaluated over a 2-week time period, and a correlation coefficient of.66 was found. Item-remainder correlations were used to assess internal consistency and ranged from.50 for useful advice, to.85 for help around the house (Broadhead, Gehlbach, DeGruy, & Kaplan, 1989). The Instrumental Support items had the poorest internal consistency. Factor analysis supported the cohesiveness of the a priori scales describing Confidant Support and Affective Support, while Instrumental Support items did not load together on a single factor. Help when I m sick in bed loaded with the Affective Support items. To improve instrument reliability the original 14-item scale was reduced to eight items. Validity Construct validity was demonstrated by significant correlations of individual items with measures of symptom status and emotional function. These measures have been shown to relate to social support. Concurrent validity was supported by significant correlations with 3 out of 4 activities measures (Broadhead, et al., 1983). Reliability and validity of the scale are supported by a study in Spain (N = 656) (Bellon Saameno, Delgado Sanchez, Luna del Castillo, & Lardilli, 1996). Factor analysis replicated the results of Broadhead and colleagues in yielding two factors Confidant Support and Affective 202
Support. Low social support was significantly related to living alone, worse subjective health, greater chronic morbidity, mental health disorder, and poorer family functioning (measured by Family APGAR). In another study using the Duke-UNC Functional Social Support Questionnaire (Williams, Williams & Griggs, 1990), social support was again shown to be strongly correlated with family functioning (measured in this instance by FACES and FCOPES). LONGSCAN Use Data Points Pre-Age 4: MW & NW sites only Age 4, 6: all sites Respondent Primary maternal caregiver Mnemonic and Version SSQB Rationale Social support received by caregivers may be an important protective factor for children at risk for maltreatment. The measure was selected because it is brief, simple to administer, and has acceptable reliability and validity. Administration and Scoring Notes LONGSCAN used a slightly modified version of the FSSQ, comprised of 10 items and 3 a priori scales: Confidant Support, Affective Support, and Instrumental Support. Seven items are from the original scale and were selected because of their demonstrated reliability and validity by the author. The other three items were developed by LONGSCAN in an attempt to enhance measurement of instrumental support. These items are: Help when I need transportation Help with cooking and housework 203
Help taking care of my children Scale scores are generated by summing the scores of all items and range from 10 to 50. Mean item scores can also be used. Results Descriptive Statistics and Reliability Table 1 presents the mean scores and standard deviations for the FSSQ at the Age 4 and Age 6 interviews by race and study site. Primary caregivers reported fairly high levels of personal social support at both times. Total mean scores across study sites were similar, with the lowest scores reported from respondents at the MW site. Table 1 about here Table 2 presents data on the scale s internal consistency reliability. Internal consistency, as measured by Cronbach s alpha, was excellent (.80-.93) across racial groups and study sites and data points. Table 2 about here Validity In order to assess validity, we correlated total scores on the FSSQ at Age 4 with Age 4 Family APGAR scores, hypothesizing that primary caregivers reporting high levels of satisfaction with family functioning would also report high amounts of personal social support. The scores on the two measures were significantly correlated (r =.25, p < 0.001). References and Bibliography Bellon Saameno, J. A., Delgado Sanchez, A., Luna del Castillo, J. D., & Lardelli C. P. (1996). Validity and reliability of the Duke-UNC 11 questionnaire of functional social support (Spanish). Atencion Primaria, 18, 153-156. Broadhead, W. E., Gehlbach, S. H., DeGruy, F. V., & Kaplan, B. H. (1988). The Duke- UNC Functional Social Support Questionnaire: Measurement of social support in family medicine patients. Medical Care, 26(7), 709-23. Broadhead, W. E., Gehlbach, S. H., DeGruy, F. V., & Kaplan, B. H. (1989). Functional 204
versus structural social support and health care utilization in a family medicine outpatient practice. Medical Care, 27(3), 221-233. Broadhead, W.E., Kaplan, B.H., James, S.A., Wagner, E.H., Schoenbach, V.J., Grimson, R., Heyden, S., & Tibblin, G. (1983). Epidemiologic evidence for a relationship between social support and health. American Journal of Epidemiology, 117(5), 521-537. Williams, P. D., Williams, A. R., & Griggs, C. (1990). Children at home on mechanical assistive devices and their families: A retrospective study. Maternal-Child Nursing Journal, 19, 297-311. 205
Table 1. Mean Scores by Race and Study Site Age 4 and 6 Interviews Age 4 Age 6 N M (SD) N M (SD) Total 1133 38.68 (8.76) 1149 38.86 (8.97) Race White 393 38.29 (8.32) 381 37.80 (8.78) Black 578 39.08 (8.81) 618 39.05 (8.87) Hispanic 81 38.33 (9.40) 85 39.98 (10.16) Multiracial 35 36.42 (11.29) 33 38.45 (9.81) Other 43 40.02 ( 7.87) 28 44.46 (5.85) Study Site EA 232 39.23 (9.01) 246 38.85 (9.37) MW 123 36.72 (8.51) 166 37.96 (9.52) SO 220 39.11 (8.93) 217 39.90 (8.77) SW 310 39.95 (8.05) 286 39.77 (8.49) NW 248 37.20 (9.05) 234 37.41 (8.72) Source. Age 4 and Age 6 data based on data received at the LONGSCAN Coordinating Center by 7/8/97 and 6/30/00, respectively. 206
Table 2. Internal Consistency Reliability (Cronbach s Alpha) by Race and Study Site Age 4 and 6 Interviews Age 4 Age 6 Total.86.87 Race White.86.87 Black.86.86 Hispanic.86.93 Multiracial.92.90 Other.82.83 Study Site EA.87.88 MW.80.88 SO.86.86 SW.86.88 NW.86.87 Source. Age 4 and Age 6 data based on data received at the LONGSCAN Coordinating Center by 7/8/97 and 6/30/00, respectively. 207