Assessing adolescents' prosocial behavior: the family helping inventory.
Within families, researchers have identified diverse patterns of interaction (Moos & Moos, 1986). For example, drawing on systems theory, Olson and McCubbin (1982) stress family cohesion and adaptability as the primary dimensions of family functioning, with effects on these dimensions coming from family processes such as support and assistance. Focusing exclusively on the siblings, Furman, Jones, Buhrmester, and Adler (1989) identified four independent dimensions of sibling relationships: warmth/closeness, relative power/status, conflict, and rivalry. The warmth/closeness dimension included helping, nurturance, intimacy, companionship, and admiration. In a similar vein, Bryant (1989) found that caretaking was composed of nurturance, challenge, punishment, and concern.
The primary methods used to study prosocial behavior among children and adolescents have been experimental analogues to socialization methods (e.g., Midlarsky, Bryan, & Brickman, 1973), home observations (Grusec, 1982; Yarrow, Scott, & Waxler, 1973), and reports by parents and teachers (Block & Block, 1973). Both for the sake of providing convergent operations, and because the child's own perceptions may be important correlates of behavior, this study was conducted to develop a self-report measure of helping within the family - the Family Helping Inventory (FHI). This paper reports the development, validation, and psychometric properties of the FHI, and of its two scales - the Sibling Helping Scale (SHS), and the Parent Helping Scale (PHS).
In defining helping behavior, a broad conceptualization was embraced. Drawing on the work by Caplan (1982), and Midlarsky & Kahana (1994); Sarason (1980), helping was defined as those activities in which one person provides physical assistance, emotional support, tangible assistance, supervision, teaching, nurturance, or general aid to another person.
Sixty-seven items were generated according to the following guidelines. First, to decrease variability attributable to subjective interpretations, items were made as behaviorally explicit as possible. Second, the items had to apply equally well to help when given to a parent or to a sibling; therefore, items which would be appropriate for only one type of family member were eliminated or rewritten. Third, every effort was made to include helping acts that could be engaged in by both boys and girls. For each item, respondents were instructed to indicate the amount of that particular kind of help he or she engaged in during the last six months for a particular sibling or for a parent (e.g., "I have tried to give comfort when this person was sad"). Responses were rated on a four-point scale, ranging from "None" to "Much."
Subjects were 202 adolescents, all of whom volunteered to participate in this study, and all of whom lived at home with their parents; 36% of the respondents were male, and 64% were female. Ages ranged from 12 to 23 years (M = 18.67, SD = 3.64). The typical respondent had three or more brothers or sisters. In addition, 20% of the sample were black, 78% were white, and 4% were Asian or other.
In addition to the Family Helping Inventory and a demographic questionnaire, respondents were administered the following battery of measures in order to measure convergent and discriminant validity.
Social Responsibility Scale (SRS) (Berkowitz & Lutterman, 1968). The goal of this scale is to assess traditional social responsibility, which is defined as an orientation toward helping others, even when there is nothing to be gained by the helper. The SRS consists of eight items, for each of which the respondent indicates the degree of agreement on a five-point scale, ranging from "strongly agree" to "strongly disagree." The internal consistency for this measure is not high (e.g., Midlarsky, Hannah, & Kahana, 1984), but its validity has been found to be satisfactory. High scorers have been found to be more altruistic than low scorers in both experimental situations (Berkowitz & Daniels, 1964; Midlarsky & Bryan, 1972), and in cross-sectional correlational studies (Midlarsky & Kahana, 1985).
Nurturance Scale of the Personality Research Form (Nurt) (Jackson, 1974). This 16-item true-false scale measures the degree to which the respondent gives comfort and sympathy to others and is caring and protective. The Nurturance Scale has been shown to have adequate reliability (.68 to .85), and to correlate with both behavioral and trait measures of nurturance (Jackson, 1974).
Self-Report Altruism Scale (SRA) (Rushton, Chrisjohn, & Fekken, 1981). On this 20-item scale, respondents indicate the frequency with which they engage in various forms of altruistic.or helping behavior. Coefficients alpha for five different samples ranged from .78 to .86. In regard to convergent validity, the SRA has been shown to correlate significantly with peer ratings of altruism, and with such behavioral measures as signing organ donor cards.
Altruistic Orientation Scale (AOS) (Midlarksy, Hannah, & Kahana, 1984). The AOS consists of eight items designed to measure the tendency to share with others, and to help others. For each item, degree of agreement is indicated on a five-point scale, ranging from "agree very much" to "disagree very much." Coefficients alpha are reported to be .83 to .90, indicating good internal consistency. Convergent validity is manifested in studies in which significant positive relationships were obtained with measures of altruistic values, self-reported helping behavior, and social responsibility (Midlarsky & Kahana, 1985).
Children's Machiavellianism Scale (Mach) (Nachamie, 1966; Christie & Geis, 1968). This 20-item scale measures the degree to which the respondent believes that people can be manipulated. Validity studies with the "Kiddie Mach" indicate that children with high scores are more likely to manipulate others in game-playing situations. Internal consistency, however, is not high, with values primarily in the .30 to .40 range, at least among the urban children studied by Nachamie (1966).
Marlowe-Crowne Social Desirability Scale (SD) (Crowne & Marlowe, 1960). This is a 33-item true-false instrument measuring the tendency to engage in impression management, and to avoid disapproval by others. Validity studies indicate that high scorers are more apt to engage in low-risk situations in order to avoid disapproval, and that they are more influenced by evaluative feedback than are low scorers. Coefficients alpha have ranged from .73 to .88, and test-retest reliability ranged from .84 to .88.
Items for the Parent Helping Scale (PHS) and the Sibling Helping Scale (SHS) were separately subjected to principal components analysis. The unrotated first factor accounted for 30.1% of the variance on the PHS and 35.9% of the variance on the SHS. Using oblique rotations, a four-factor solution for the SHS, and a five-factor solution for the PHS yielded the most meaningful factor structures. Items were included on a factor only if the factor loading was .40 or above. Items were excluded if they loaded above .40 on two factors or about .40 on one factor, and above .30 on two other factors. Given these criteria, 19 items were dropped from the SHS and 15 items were dropped from the PHS. In most cases, subscales comprising the same items were given the same name on both the SHS and the PHS.
For the SHS, the following four factors were retained: Emotional Support, Custodial Care, Information-Giving, and Tangible Aid. The five factors for the PHS are: Emotional Support, Information-Giving, Custodial Care, and Love/Affection. Reliabilities for each subscale and for the total scales were calculated using coefficients alpha. The alphas ranged from .70 to .96, indicating a good level of internal consistency. Scores for respondents were obtained by summing the unweighted raw scores for all items on the respective subscale. For each subscale, the factors and their eigenvalues, coefficients alpha, means and standard deviations are presented in Table 1.
Convergent and Discriminant Validity
In order to assess convergent validity, the relationships of the PHS, the SHS, and their subscales to the other self-report measures of altruism and helping were first determined. Discriminant validity was investigated by determining the pattern of relationships of the scales and subscales of the Family Helping Inventory (FHI) to the measures of Machiavellianism and of social desirability. Zero-order correlations among the measures, and the correlation coefficients corrected for attenuation of measurement (i.e., unreliability) are presented in Table 2.
Table 1 Factor Eigenvalues, Means, Standard Deviations, and Coefficients Alpha for the PHS and the SHS Coefficient Standard Eigenvalue Alpha Mean Deviation PHS Emotional Support 30.1 .93 45.22 9.97 Information 6.5 .91 34.67 8.08 Tangible Aid 4.4 .81 36.70 5.77 Love 3.4 .93 31.06 7.40 Custodial Care 2.8 .70 6.95 2.76 Total .97 191.44 33.91 SHS Emotional Support 35.9 .96 75.71 15.85 Custodial Care 6.1 .80 17.16 5.00 Information 4.7 .92 41.95 9.18 Tangible Aid 3.1 .79 19.87 4.67 Total .97 202.61 33.91
Evidence for convergent validity comes from the finding of significant positive relationships between the pre-existing measures of prosocial tendency and the Total Helping scores of the PHS and the SHS (and of the preponderance of the PHS and SHS subscales.) For both scales, therefore, findings indicated a consistent pattern of relationships with measures of social responsibility (Berkowitz & Lutterman, 1968), nurturance (Jackson, 1974), altruistic orientation (Midlarsky, Kahana, & Hannah, 1984), and self-reported helping of strangers (Rushton, Chrisjohn, & Fekken, 1981). In addition, the SHS predicted a composite of the four other measures at r = .49, p [less than] .001. This correlation rose to r = .56, p [less than] .001 when corrected for attenuation, and remained unchanged when the effects of social desirability were partialed out. The PHS predicted the same composite at r = .42, p [less than] .001 when corrected for attenuation. After covarying the effects of social desirability, the relationship between the PHS Total Helping score and prosocial tendency became r = .43, p [less than] .001.
In considering discriminant validity, supportive evidence comes from the finding of a pattern of nonsignificant relationships, and of significant inverse relationships with Machiavellianism. Regarding social desirability, for the SHS only the Total Helping score had a significant positive relationship. On the other hand, the PHS Total Helping score, and scores on four of the PHS subscales were related to social desirability (see Table 2). This set of relationships raised the question of the degree to which scores on the PHS in particular - and on the FHI in general - may reflect evaluative dependency (i.e., approval motivation) rather than prosocial motivation. Hence, partial correlations of the PHS, the SHS, and their subscales with all of the self-report measures of prosocial tendencies were calculated, holding social desirability constant. These results are presented in Table 3. A comparison of the results in Tables 2 and 3 reveals that the relationships between the measures of prosocial tendencies and the components of the Family Helping Inventory remained virtually unchanged when the effects of social responsibility were removed.
As a further check on convergent validity, we determined the degree to which the PHS and the SHS were related to self-reported helping behaviors. These behavioral measures were selected because they have been used to validate previous helping scales (Rushton, 1984). As Table 4 shows, 32 of the 44 correlations between FHI subscales and the [TABULAR DATA FOR TABLE 2 OMITTED] behavioral measures were significant at p [less than] .05 or better. In addition, the PHS predicted a composite of the four behavioral measures at r = .18, p [less than] .05, and r = .21, p [less than] .01 when corrected for attenuation. The SHS predicted the same composite at r = .25, p [less than] .001, which was still r = .25, p [less than] .001 when corrected for attenuation.
Table 3 Partial Correlations of the PHS and SHS with Other Self-Report Scales of Prosocial Tendencies, Holding Social Desirability Constant Social Nurturance Self-Report Altruistic Respons. Altruism Orientation PHS Emotional Support [.23.sup. =] .47(a) .39(a) .33(a) Information .07 .24(a) .34(a) .14(c) Tangible Aid .23(a) .43(a) .55(a) .32(a) Love/Affect .22(a) .41(a) .31(a) .32(a) Cust. Core -.21(b) .27(a) .28(a) .02 Total .20(b) .46(a) .46(a) .32(a) SHS Emotional Support .35(a) .52(a) .44(a) .32(a) Cust. Care .03 .37(a) .30(a) .11 Information .38(a) .40(a) .46(a) .27(a) Tangible Aid .22(a) .43(a) .55(a) .20(b) Total .33(a) .51(a) .49(a) .28(c) a p [less than] .001 b [less than] .01 c [less than] .05 Table 4 Correlation of the PHS and SHS with Behavioral Indices and a Self-Report Rating of Helping Organ Amt. Give Learn Self- Donor Vol. Blood To Help Rating PHS Emo. Sup. .08 .13(c) .07 .10 .29(a) Informa. -.07 .05 .05 .14(c) .21(b) Tang. Aid .01 .21(b) .19(b) .19(b) .25(a) Love/Aff. .12(c) .16 .16 .16(c) .20(b) Cust Care -.14(c) -.02 .03 .16(c) .20(b) Total .20 .16(c) .15(b) .13(c) .30(a) SHS Emo. Sup. .27(a) .28(a) .19(b) .14(c) .30(a) Cust.Care .08 .22(a) .12(c) .09 .16(c) Informa. .16(c) .27(a) .19(b) .21(b) .22(a) Tang Aid .08 .18(c) .15(c) .15(c) .30(a) Total .20(b) .27(a) .21(b) .21(b) .32(a) a p [less than] .001 b p [less than] .01 c p [less than] .05
Sex and Age Differences
To determine whether boys and girls are differentially helpful, MANOVAS were conducted on the subscales, and ANOVAS on the total scale scores of both the PHS and the SHS. The MANOVA on the subscales, using Wilks' lambda criteria, indicated a significant effect for gender for both the PHS, F(5, 179) = 5.51, p [less than] .001, and the SHS, F(5, 144) = 6.12, p [less than] .001. Subsequently, univariate one-way ANOVAS were calculated on the subscales (see Table 5). For the PHS, girls scored significantly higher on Emotional Support, Tangible Aid, Love/Affection, and Custodial Care than did boys. For the SHS, girls scored significantly higher on Emotional Support, Custodial Care, and Information than did boys. ANOVAS calculated on the Total Scores of the PHS indicated that females engaged in significantly more help than did males, F(1, 191) = 14.62, p [less than] .001. Similar results were obtained for the SHS, F(1, 151) = 14.93, p [less than] .001.
In addition, PHS and the SHS scores were significantly related to the age of the respondent. As Table 6 shows, on the PHS, age was positively correlated with Information, Tangible Aid and Custodial Care, as well as with Total Helping. For the SHS, Tangible Aid was positively related to age, and Custodial Care was negatively related.
Since children and adolescents have different kinds of relationships with their brothers and sisters than they do with their parents, it is not surprising that there is some variation in the kinds of help they provide to each family member. Results of the separate factor analyses indicated enough similarity in the items loading on specific factors to consider them to be equivalent activities. Thus, children and adolescents provide the following kinds of help to both siblings and parents: Information, Tangible Aid, and Custodial Care. While Emotional Support was also provided to both groups, for parents the items split into two factors - Emotional Support and Love/Affection. In contrast, for siblings the items loaded only on the Emotional Support factor.
Further, the analyses performed thus far indicate that both scales of the FHI have good psychometric properties. In regard to convergent validity, each FHI subscale correlates more strongly with its corresponding measure than it does with any other measure. For example, Emotional Support and Love/Affection have their highest correlations with the Nurturance Scale of the Personality Research Form (Jackson, 1974), while Tangible Aid has its highest-correlation with the Self-Report Altruism Scale (which primarily asks about instrumental helping) (Rushton, Chrisjohn, & Fekken, 1981).
Table 5 F-Tests and Means for Gender Differences on the PHS and SHS Mean Males Females F Value PHS Emo. Sup. 42.15 (9.05) 47.18 (9.98) 11.59(a) Informa. 33.92 (8.41) 35.05 (9.11) .70 Tang. Aid 35.38 (5.07) 37.83 (5.98) 7.97(b) Love/Aff. 28.43 (7.10) 32.81 (7.10) 16.06(a) Cust. Care 6.12 (2.22) 7.49 (2.95) 10.94(a) Total 179.93 (30.33) 198.64 (34.08) 14.62(a) SHS Emo. Sup. 68.95 (15.79) 80.36 (14.33) 20.48(a) Cust. Care 15.23 (4.74) 18.33 (4.79) 14.70(a) Informa. 39.90 (9.42) 43.40 (8.81) 5.23(c) Tang. Aid 19.04 (4.62) 20.46 (4.62) 3.29 Total 188.23 (36.99) 211.36 (34.87) 14.93(b) a p [less than] .001 b p [less than] .01 c p [less than] .05 Table 6 Correlations of the PHS and SHS with Child Age Age PHS Emotional Support .03 Information .21(b) Tangible Aid .23(b) Love/Affection .07 Custodial Care .12(c) Total .19(b) SHS Emotional Support -.03 Custodial Care -.17(c) Information .02 Tangible Aid .16(c) Total -.02 b p [less than] .01 c p[less than] .05
All subscales also correlated moderately with both the Social Responsibility Scale (Berkowitz & Lutterman, 1968) and the Altruistic Orientation Scale (Midlarsky, Kahana, & Hannah, 1984), suggesting that helping within the family is also related to a general predisposition to assist others; significant relationships with behavioral self-reports of helping outside the family context were significant as well. In addition, and demonstrating discriminant validity, the scales did not correlate, or correlated negatively with the Children's Machiavellianism Scale (Christie & Geis, 1968; Nachamie, 1966).
While several subscales had significant positive relationships with social desirability (Crowne & Marlowe, 1960), when this variable was covaried, the pattern of overall relationships remained virtually unchanged. Thus, it appears that the FHI has both convergent and discriminant validity. In addition, reliability as determined by coefficients alpha is more than adequate.
In regard to sex differences, girls reported engaging in significantly more Emotional Support, Tangible Aid, Love/Affection, Custodial Care, and Total Help to parents than did boys. Girls also reported giving significantly more Emotional Support, Custodial Care, Information, and Total Help to siblings than did boys. Such a finding supports the notion that helping within families is in fact a task more often relegated to females. On the PHS, age was positively and significantly correlated with Information, Tangible Aid, Custodial Care, and Total Help. On the SHS, age was positively related to Tangible Aid. Taken together, the pattern of results suggests that as children grow older, they provide increasing amounts of instrumental help to both parents and siblings. It is probable that this age-related trend is associated with greater intellectual and physical competence (Midlarsky, 1984), as well as to a wide range of experiences.
In sum, then, the Family Helping Inventory seems to be a promising instrument for assessing self-reported helping behavior directed by children toward a sibling or parent. It should prove useful to researchers interested in studying processes that occur in families, e.g., single-parent, handicapped member. Future research should be directed toward confirmatory factor analyses, further establishing validity with direct observations of helping behavior, and additional assessments of reliability, particularly test-retest reliability.
Bar-Tal D. Raviv, A. B. & Goldberg, M. (1982). Helping behavior among preschool children: An observational study. Child Development, 53, 396-402.
Berkowitz, L., & Daniels, L. R. (1967). Responsibility and dependency. Journal of Abnormal and Social Psychology, 66, 429-436.
Berkowitz, L., & Lutterman, K. (1968). The traditionally socially responsible personality. Public Opinion Quarterly, 32, 169-185.
Berman, P. W. (1987). Children caring for babies: Age and sex differences in response to infant signals and to the social context. In N. Eisenberg (Ed.), Contemporary topics in developmental psychology (pp. 141-164). New York: Wiley.
Block, J. H. (1973). Conceptions of sex roles: Some cultural and longitudinal perspectives. American Psychologist, 28, 512-526.
Block, J., & Block, J. H. (1973). Ego development and the provenance of thought. Progress report for NIMH. Cited in N. Eisenberg (Ed.) (1982). The development of prosocial behavior. New York: Academic Press.
Bryant, B. K. (1989). The child's perspective of sibling caretaking and its relevance to understanding social-emotional functioning and development. In P. Zuckow (Ed.), Sibling interaction across cultures. New York: Springer-Verlag.
Caplan, G. (1982). The family as a support system. In H. I. McCubbin, A. E. Cauble, & J. M. Patterson (Eds.), Family stress, coping and social support (pp. 200-220). Springfield: Charles C. Thomas.
Christie, R., & Geis, G. (Eds.). (1968). Studies in Machiavellianism. New York: Academic Press.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354.
Eisenberg, N., & Lennon, R. (1980). Altruism and the assessment of empathy in the preschool years. Child Development, 51, 552-557.
Eisenberg, N., & Mussen, P. H. (1989). The roots of prosocial behavior in children. New York: Wiley.
Furman, W., Jones, W., Buhrmester, D., & Adler, T., (1989). Children's, parents', and observers' perspectives on sibling relationships. In P. Zuckow (Ed.), Siblings interactions across cultures. New York: Springer-Verlag.
Gottman, J. M., & Parkhurst, J. T. (1980). A developmental theory of friendship and acquaintanceship processes. In W. A. Collins (Ed.), Development of cognition, affect, and social relations. The Minnesota symposium on child psychology. (Vol. 13, pp. 197-153). Hillsdale, NJ: Erlbaum.
Grusec, J. E. (1982). The socialization of altruism. In N. Eisenberg (Ed.), The development of prosocial behavior. New York: Academic Press.
Jackson, D. N. (1974). Personality Research Form Manual. Port Huron, MI: Research Psychologists Press.
Martin, G. B., & Clark, R. D. III (1982). Distress crying in neonates: Species and peer specificity. Developmental Psychology, 18, 3-9.
Midlarsky, E. (1984). Competence and helping: Notes toward a model. In E. Staub, D. Bar-Tal, J. Karylowski, & J. Reykowski (Eds.), Development and maintenance of prosocial behavior (pp. 291-308). New York: Plenum Press.
Midlarsky, E., & Bryan, J. H. (1972). Affect expressions and children's imitative altruism. Journal of Experimental Research on Personality, 6, 195-203.
Midlarsky, E., Bryan, J. H., & Brickman, P. (1973). Aversive Approval: Interactive effects of modeling and reinforcement on altruistic behavior. Child Development, 44, 321-328.
Midlarsky, E., & Hannah, M. E. (1985). Competence, reticence, and helping among children and adolescents. Developmental Psychology, 21, 534-591.
Midlarsky, E., & Hannah, M. E. (1989). The generous elderly: Naturalistic studies of donations across the life span. Psychology and Aging, 4, 346-351.
Midlarsky, E., Hannah, M. E., & Kahana, E. (1984). The Altruism Orientation Scale. Working paper, Center for the Study of Development and Aging, University of Detroit.
Midlarsky, E., & Kahana, M. E. (1985). Personal and situational influences on late life helping. Humboldt Journal of Social Relations, 13, 217-233.
Moos, R. H., & Moos, B. S. (1986). Family Environmental Scale Manual. Palo Alto, CA: Consulting Psychologists Press.
Nachamie, S. (1966). Machiavellianism in children: Construction and validation of the Children's Mach Scale. American Psychologist, 21, 628.
Olson, D. H., & McCubbin, H. I. (1982). Circumplex model of marital and family systems, V. Applications to family stress and crisis intervention. In H. I. McCubbin, A. E. Cauble, & J. M. Patterson (Eds.), Family stress, coping, and social support. Springfield: Charles C. Thomas.
Radke-Yarrow, M., Zahn-Waxler, C., & Chapman, M. (1983). Prosocial dispositions and behavior. In P. Mussen (Ed.), Manual of child psychology. Vol. 4; Socialization, personality, and social development (pp. 469-545). New York: Wiley.
Rushton, J. P. (1984). The altruistic personality. In E. Staub, D. Bar-Tal, J. Karylowski, & J. Reykowski (Eds.), Development and maintenance of prosocial behavior (pp. 271-290). New York: Plenum.
Rushton, J. P., Chrisjoh, R. D., & Fekken, G. C. (1981). The altruistic personality and the Self-Report Altruism Scale. Personality and Individual Differences, 2, 293-302.
Sarason, I. G. (1980). Life stress, self-preoccupation and social support, In I. G. Sarason, & D. D. Spielberger (Eds.). Stress and anxiety, Vol. 7. Washington, DC: Halsted.
Whiting, B, & Edward, C. P. (1973). A cross-cultural analysis of sex differences in the behavior of children aged three through eleven. Journal of Social Psychology, 91, 171-188.
Yarrow, M. R., Scott, P. M., & Waxler, C. A. (1973). Learning concern for others. Developmental Psychology, 8, 240-260.
Zahn-Waxler, C., Radke-Yarrow, M., & King, R. (1983). Early altruism and guilt. Academic Psychology Bulletin, 5, 247-260.
|Printer friendly Cite/link Email Feedback|
|Author:||Midlarsky, Elizabeth; Hannah, Mary Elizabeth; Corley, Robin|
|Date:||Mar 22, 1995|
|Previous Article:||Adolescents' intimacy with parents and friends.|
|Next Article:||Family environments of adolescent sex offenders and other juvenile delinquents.|