|
Abstract
|
This paper presents a shortened, 20-item version of The Parental Bonding Instrument (PBI), used in a sample of 573 Norwegian adolescents aged 15-19. The instrument measures mothers' and fathers' "care" and "control/overprotection" as perceived by the adolescent. The two dimensions of the instrument were clearly defined using factor analyses. In addition, they were shown to be related to measures of both depression/anxiety and delinquency, showing that PBI scores may indicate general and nonspecific risk factors in a broad range of psychosocial problems.
VioLit summary:
OBJECTIVE:
The aim of this study by Pedersen was to present The Parental Bonding Instrument (PBI), a measure of adolescents' perceptions of their mothers' and fathers' "care" and "control/ overprotection."
METHODOLOGY:
The author conducted a primary analysis of quasi-experimental data collected from 267 male and 306 female respondents participating in a longitudinal study in Oslo, Norway, "Youth, Lifestyle, and Drugs." (The author stated that for details of design, selection, and methods used in this study, the reader should refer to Pedersen, 1990; 1991). Respondents were aged 15-19 years. Data were collected during school hours in fall, 1987, and then in spring, 1988. The author reported that 97.2% of participants responded to their questionnaire in 1987; 95.8% responded in 1988. The author chose 10 items from the Parental Bonding Instrument developed by Parker, Tupling, and Brown, 1979. This instrument was designed to measure the levels of "care" and "control/overprotection" provided by parents to their children. The questionnaire elicits responses from adult respondents regarding perceptions of their parents as they were growing up (up to 16 years old). Five items measured a "care" dimension (e.g., "They have been affectionate toward me"), and five items measured a "control/overprotection" dimension (e.g., "They had allowed me to decide things myself"). The author reported that a high care score indicated empathetic, warm, understanding, and friendly parents. A low care score intimated a perception of cold, unhelpful, rejecting and insufficiently caring, parents. A high control/overprotection scale score indicated parents who were overprotective, caused the child to stay infantile, were controlling and invasive. A score at the opposite end of this scale indicated that the child was allowed autonomy. The author stated that the items were chosen according to how well they were understood by the students. The revised PBI instrument was administered in 1987 only. As a measure of anxiety, depression, and level of social function, the respondents were also requested to complete the General Health Questionnaire 12 (GHQ 12). Levels of delinquency were measured using six questions associated with activities such as school truancy, shoplifting, and police contact. The author stated that values ranged from 0-19. The latter two instruments were presented in 1988.
FINDINGS/DISCUSSION:
The author found that alpha values for the four subscales ranged from .69 to .78. A correlation coefficient of .49 was found between mother's and father's care; .48 between mother's and father's control. The correlation coefficient between care and control for father was -.33, and -.44 for mother. The author conducted a varimax-rotated factor analysis of the PBI scores for both mother and father. It was argued that a two-factor solution was most suited to the data. The author found that care was the primary and overriding factor, followed by overprotection, for fathers. The opposite was discovered for mothers. The author reported that perceptions of father did not significantly differ between males and females. However, according to the female respondents, mothers appeared to be more caring and less overprotective. Male respondents perceived their mother to be more controlling than their father (t=-.46, df=238, p=0.000). Mothers were considered significantly more caring than the father by both males and females (t=12.6, df=493, p=0.000). The General Health Questionnaire 12 revealed that 12% of males and 21% of females scored in the symptom area of poor mental health. The author found that 11% of females, and 23% of males fell into the delinquency range. Approximately 7% of both sexes had both poor mental health problems and delinquency.
The author conducted a further 4 ANOVAS on lower and higher cutoff points for both the "care" and "overprotection/care" measures. They obtained similar results in which a reversed relationship between delinquency and poor mental health was found: girls showed anxiety/depression symptoms and boys showed delinquency symptoms. The author argued that the findings were in line with previous research: females tended to internalize and males were inclined to externalize psychosocial problems. The author found that the "no symptoms" group reported high care and low control, where as the "anxiety/depression" group reported higher control and lower care when compared to the "no symptoms" group. The author reported that, compared with the other three groups the symptom free group reported significantly higher levels of father's care (F(1,3)=12.0, p=0.000), and mother's care, (F(1,3)=9.7, p=0.000). Father's control was higher for the group experiencing both anxiety/depression and delinquency (both conditions), (F(1,3)=4.6, p=0.003). The group with both conditions and the delinquency only group reported higher levels of mother's control than did the group without symptoms (F(1,3)=6.3, p=0.000).
A number of stepwise canonical discriminant analyses were conducted. The author wrote that parents' care and overprotection/control, together with sex, were utilized to differentiate between the symptom groups and symptom free group. It was found that sex made a difference with regards anxiety/depression and delinquency, but didn't make a difference when both problems featured. Father's and mother's care were found to be the most discriminating variables: low paternal care was most characteristic of the group exhibiting anxiety/depression when compared with the symptom free group; low maternal care characterized the delinquency group and the group exhibiting both symptoms.
The author concluded that there were methodological problems associated with the current study. These included: 1) the possibility that the PBI instrument may be affected by personality or mood; 2) the questionnaires were administered to adolescents who may have been in current conflict with their parents which may have biased results; 3) the degree of severity of anxiety, depression and/or delinquency may not have been adequately differentiated since wide measures of these variables were used; and 4) anxiety/depression and delinquency may be related, rather than dissimilar phenomena, therefore future studies should assess the association between the two. The author argued that, even so, the findings had demonstrated that the PBI possessed adequate psychometric properties.
AUTHOR'S RECOMMENDATIONS:
The author articulated that, based on these methodological flaws, caution before drawing conclusions was advisable. Even so, it was argued that the findings revealed a relationship between anxiety/depression and/or delinquency, and levels of parental care and control. Low paternal care was the strongest predictor of anxiety/depression, and low maternal care was the strongest indicator of delinquency, and anxiety/depression and delinquency combined. Based on the findings future use of the Parental Bonding Instrument with adolescents was recommended.
(CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado)
1980s
Countries Other Than USA
Norway
Interparental Relations
Interparental Conflict
Couple Relations
Couple Conflict
Family Conflict
Family Relations
Juvenile Perceptions
Survey Instrument
Instrument Description
Juvenile Delinquency
Juvenile Adjustment
Juvenile Mental Health
Delinquency Causes
Juvenile Offender
Language: en |