Financial stress, social support, and alcohol involvement: A longitudinal test of the buffering hypothesis in a general population survey.
Thoughts
Head empty
Annotations
peirce1996 - p. 39
The second way in which social support was expected to be related to alcohol involvement was that it may moderate, or buffer, the relationship between financial stress and alcohol involvement. The buffering hypothesis asserts that social support protects (i.e., buffers) people from the deleterious effects of stress on alcohol use.
peirce1996 - p. 39
Respondents in this study were drawn from a random sample survey of household residents in Erie County, New York, who took part in a longitudinal study of stress processes.
peirce1996 - p. 40
Two measures of financial stress were assessed at Wave 1. First, chronic financial stress was assessed with a five-item measure developed by Pearlin and colleagues (Pearlin & Lieberman, 1979; Pearlin et al., 1981). Four items assessed how often the respondent did not have enough money to afford food, medical care, clothing, and family leisure activities. The response anchors ranged from almost never-never (1) to almost always (4). The fifth item assessed how the respondents’ finances usually worked out at the end of the month. The response anchors ranged from some money left over (1) to not enough money to make ends meet (3).
peirce1996 - p. 40
Second, a measure of acute financial stress was derived from financial life events taken from the Psychiatric Epidemiologic Research Inventory Life Events Scale (Dohrenwend, Krasnoff, Askenasy, & Dohrenwend, 1978). There were five specific events: having less money than usual, having to borrow money, going on welfare, experiencing a foreclosure on a loan or mortgage, and reporting a miscellaneous event. In addition to reporting whether each event occurred during the preceding 12months, respondents were asked to rate the overall effect of the event on their lives using a 6-point scale ranging from extremely positive (1) to extremely negative (6).
peirce1996 - p. 40
erceived social support. Three social support subscales were assessed at Wave 1 from a set of 30 items (the 10 items assessing self-esteem support were not used) from the Interpersonal Support Evaluation List (Cohen, Mermelstein, Kamarck, & Hoberman, 1985). The 15 items chosen for this study were based on previous factor analyses and were found to be the 5 best indicators for each subscale (see the Appendix for the items used in this study).
peirce1996 - p. 40
lcohol involvement. Three dimensions of alcohol involvement were assessed at both Wave 1 and Wave 2. First, drinking to cope was assessed with a 5-item scale (Cooper, Russell, Skinner, & Windle, 1992) that required respondents to report the frequency of drinking to manage or cope with negative emotions (e.g., “to forget your worries” and “to cheer up when you’re in a bad mood”). A 4-point response scale was used, with options ranging from almost never-never (1) to almost always (4). An overall drinking to cope score was created by averaging responses to all 5 items.
peirce1996 - p. 40
frequency of heavy drinking was assessed by askingrespondents how often they had consumed five or more drinks in a single day duringthe past 12months. The response anchors ranged from never (0) to five times a week or more (8).
peirce1996 - p. 40
alcohol problems were assessed with 17 items designed to yield a Diagnostic and Statistical Manual of Mental Disorders (3rd edition; American Psychiatric Association, 1980) diagnosis of alcohol abuse and dependence; these items were taken from the National Institute of Mental Health Diagnostic Interview Schedule (Robins, Helzer, Croughan, Williams, & Spitzer, 1981).
peirce1996 - p. 40
The following six sociodemographic covariates were assessed at Wave 1: sex (0 = male, 1 = female), age (in years), race (0 = non-Black, 1 = Black), marital status (0 = not married, 1 = married or living as though married), family income (annual dollar amount), and family education (in years).
peirce1996 - p. 42
T-H
peirce1996 - p. 42
S