|
This is part 2 from Intrapersonal and Organizational Factors Associated With Burnout Among School Counselors ParticipantsParticipants were drawn from the full membership list of a school counseling organization in the northeastern United States (N= 611). We used a systematic random sample with a random start. The membership list, organized across 10 different regions, yielded a proportionally representative sample of school counselors (N= 202) from each region. Ninety-four surveys were returned, resulting in an overall response rate of 47%. Sixteen of the 94 respondents had either retired or were currently employed in other positions (e.g., counselor education, social work) and hence were excluded from the data analysis stage. The final sample comprised 22 men (28%) and 56 women (72%) (n = 78) representing a 39% response rate.
Participants ranged in age from 24 to 70 years old (M = 43.19, SD = 11.4). Seventy (89.7%) participants identified themselves as White, 3 (3.8%) as Black or African American, 2 (2.6%) as American Indian or Alaska Native, 2 (2.6%) as Hispanic or Latino/a, and 1 (1.3%) as Asian/Asian American or Pacific Islander. Eighteen (23.1%) respondents were single, and 50 (64.1%) were either married or involved in a same-sex partnership. Nine (11.6%) were separated, widowed, or divorced (1 person did not report partnership status). In terms of education, 43 (55.1 %) participants earned their degrees at graduate programs accredited by the Council for Accreditation of Counseling and Related Educational Programs. Fifty-five (70.5%) respondents indicated master's degrees as their highest level of education attained, 18 (23.1%) reported certificates of advanced studies, and 4 (5.1%) reported doctoral degrees (one person did not report this demographic). Overall, respondents had approximately 11 years of experience in the profession (M =11.04, SD = 8.2) and had been working in their current positions for about 7 years (M= 7.19, SD = 6.5). Forty-three (55.1%) counselors reported employment at the high school level, 16 (20.5%) at the middle school level, and 7 (9.0%) at the elementary school level, while 12 (15.4%) others reported working in an alternatively configured environment (e.g., Grades 7-12). For this particular sample, 70 (89.7%) respondents worked in public schools, and 8 (10.3%) worked in independent or other types of schools. Forty-three (55.1%) counselors presently worked in suburban schools, 20 (25.6%) worked in rural schools, and 14 (17.9%) were currently employed in urban settings (1 person did not report this demographic).
MeasuresBased on the research design, data for three different sets of predictor variables were gathered for this study.
Demographic questionnaire. As noted, there is research to suggest correlations between certain demographic variables and burnout. For this study, the four most salient dimensions from the literature were included in the analysis: age, student/counselor ratio (SCR), total years of school counseling experience (YOE), and whether or not counselors were receiving some form of regular clinical supervision (SUP). These predictor variables were entered first in the data analysis because they represent background characteristics that may be associated with burnout but are less amenable to change than subsequent variables entered.
Coping Inventory/or Stressful Situations (CISS). The CISS (Endler & Parker, 1999) was used to assess the intrapersonal dimension. This 48-item inventory uses a 5-point Likert-type scale and measures three main strategies for coping with stress. The Task-Oriented Coping (TOC) scale measures the extent to which individuals deal directly with problems at hand. The Emotion-Oriented Coping (EOC) scale measures the extent to which people concentrate primarily on the resulting emotions brought on by stressful situations. Finally, the Avoidance-Oriented Coping (AOC) scale measures the extent to which people try to avoid existing stressful situations in their lives (Cosway, Endler, Sadler, & Dreary, 2000). Respondents are asked to indicate how much they engage in a variety of activities (e.g., "scheduling my time better") when they encounter difficult, stressful, or upsetting situations. The responses range from 1 (not at all) to 5 (very much). The inventory has been normed on a number of different populations including adults (N= 537), undergraduates (N = 1,242), psychiatric inpatients (N = 302), and early (N= 313) and late adolescents (N= 504; Endler & Parker, 1999). Construct validity has been examined and upheld in numerous studies. Endler and Parker also reported highly satisfactory alpha coefficients on each factor across the different normative groups ranging from .87 to .92 on the TOC scale, .82 to .90 on the EOC scale, and .76 to .85 on the AOC scale. Alpha coefficients for the CISS in the current study were also satisfactory (TOC, a= .85; EOC, a =.90; and AOC, a =.80).
Two inventories were used to collect data on organizational factors associated with burnout.
Role Questionnaire. To determine how individuals perceived their work environments, participants were asked to fill out the Role Questionnaire (RQ; Rizzo, House, & Lirtzman, 1970). This 14-item questionnaire was developed to measure the degrees of role conflict and role ambiguity that an individual experiences on the job. Each of the items on this survey is rated on a 7-point Likert scale. Participants are asked to respond to questions such as "I have to work on unnecessary things" and to rate these statements on a scale from 1 (very untrue) to 7 (very true). A score of 1 indicates strong agreement that the item reflects the individual's job situation, whereas a score of 7 indicates strong disagreement with the item. Schwab, Iwanicki, and Pierson (1983) reported Cronbach's alpha reliabilities of .85 for the Role Conflict (RC) scale and .86 for the Role Ambiguity (RA) scale when used with teachers. Validity and reliability have also been upheld in studies among high school supervisors, special education teachers, manufacturing supervisors, foremen, salespersons, clerical staff, nurses, public utility workers, hospital staff, and hospital aides (Kottcamp & Mansfield;Rizzo et al.; Schuler, Aldag, & Brief; Van Sell, Brief, & Schuler; all as cited in Freeman & Coll, 1997).
Freeman and Coll (1997) performed a factor analysis on the RQ with high school counselors (-/V= 417) and found it to measure three dimensions. They added Role Incongruity (RI) to their factor solution and defined this term as the difference between the roles a school counselor actually performs and the roles he or she believes ought to be performed. Because the current study focuses on school counselors, Freeman and Coil's three-factor model was used. Higher scores on the RI and RC scales indicate higher degrees of job incongruity and conflict. Conversely, it is lower scores on the RA scale that indicate increased levels of job ambiguity. Alpha coefficients for the RC, RI, and RA scales in the current study were highly satisfactory (RC, <x= .75;RA,a=.90;RI,a=.85)
Counselor Occupational Stress Inventory (COSI; Gray, 1982). To further assess school counselor organizational factors associated with burnout, specific sections of the COSI were used. The COSI was designed specifically for use with school counselors. On this 50-item inventory, participants are asked to respond to typical school counselor situations by assessing the level of stress they associate with each situation. For example, on the original COSI, a situation (e.g., "Sitting in on teacher-parent or teacher-student conferences") is presented and respondents are asked to rate this on a 5-point Likert scale with response possibilities ranging from 1 (not stressful) to 5 (very stressful). The response format for this inventory, however, is sometimes awkward. For that reason, the original COSI was modified slightly for this study. All items were converted to sentence stems. For example, the situation just mentioned was changed to: "Sitting in on teacher-parent or teacher-student conferences is: ... ." Respondents were given the same 5-point Likert scale ranging from not stressful to very stressful. We believe that this slight modification helped to alleviate potential respondent confusion.
Factor analysis of the COSI yielded six subscales to measure job stress: Financial Security (FS), Lack of Decision-Making Authority (LDMA), Counselor-Teacher Professional Relationships (CTPR), Counselor-Principal Professional Relationships (CPPR), Non-Professional Duties (NPD), and Professional Job Overload (PJO). For this particular study, we used only four of the developed stress subscales from the COSI (FS, LDMA, CTPR, CPPR). Visual inspection of the two other subscales, NPD and PJO, revealed apparent overlaps with the three subscales already measured by the RQ and, thus, were excluded from the survey. Using only four subscales from the COSI as opposed to six reduced the number of items for this section of the survey from 50 to 29. Gray (1982) reported internal consistency scores on the subscales of the COSI ranging from 0.81 to 0.95. This study yielded similar results (FS, a- .97; LDMA, a= .93; CTPR, a= .92;and CPPR, a =.89)
MBI (Maslach, Jackson, & Leiter, 1996). The outcome variable, burnout, was assessed using the MBI. This widely used, 22-item inventory has three different versions: the MBI-Human Services Survey (MBI-HSS), the MBI-Educators Survey (MBI-ES), and the MBI-General Survey (MBI-GS). The MBI-ES was designed specifically for use with the education population and thus was chosen for this particular study. The only difference between it and the MBI-HSS is the substitution of the word students for recipients in one third of the items (Sandoval, 1989). This inventory uses a 7-point Likert scale. For each item, respondents are asked to indicate how often a particular feeling or attitude has been experienced in their work environment. The responses range from 0 (never) to 6 (every day). The inventory yields results on three different subscales: Emotional Exhaustion (EE), Depersonalization (DP), and Personal Accomplishment (PA). There is no summed or total score for the entire survey (Hargrove, 1989). For the MBI-ES, high degrees of burnout are indicated by elevated scores on the EE and DP subscales and depressed scores on the PA subscale. Iwanicki and Schwab (1981) reported Cronbach's alpha coefficients for the three different subscales on the MBI-ES as .90 for EE, .76 for DP, and .76 for PA. Gold (1984) reported estimates of .88, .74, and .72, respectively. Maslach et al. (1996) noted that these reliability coefficients closely parallel those of the MBI-HSS, the more frequently used version of the MBI. For the current study, reliability analysis yielded similar results (EE, a = .91;DP, a= .74; and PA, oc= .78). Huberty and Huebner (1988) validated the inventory's factor structure for school psychologists, but to date there have been no validation studies among school counselors.
ProcedurePermission to reproduce copyrighted inventories onto one single document was secured. Both copyrighted inventories (the CISS and the MBI-ES) were reproduced in their original format. The survey packet, including the COSI (Gray, 1982), RQ (Rizzo et al, 1970), CISS (Endler & Parker, 1999), and MBI-ES (Maslach et al., 1996), contained 123 items plus a demographics section. Participants also received a consent form explaining the nature of the study and soliciting their signed, informed consent to participate. A first-time mailing (N = 202) was conducted on January 24, 2003. Follow-up postcards were sent to nonrespondents 19 days later.
Data Entry and AnalysisData were entered as surveys were returned. Regarding responses to items using the Likert scale, in all cases where respondents chose to circle two numbers instead of one, the lower of the two numbers was entered into the data file. In the event of skipped questions, mean scores were inserted. Survey responses of "not applicable" were entered into the data file at the lowest response set level (either 0 or 1 depending on the item). Overall, respondents answered 99% of the questions on their surveys.
SPSS 11.5 was used to analyze the data. A post hoc power analysis was conducted using G*Power© (Buchner, Faul, & Erdfelder, 1997). In addition to generating descriptive statistics, a hierarchical regression was completed to determine both the separate and the combined contributions that each independent variable set made to the outcome measures on the MBI-ES. The analysis also identified the most influential individual variables in predicting the outcome in the final models. In accordance with hierarchical regression procedures, the order in which independent variable sets were entered was determined a priori (Licht, 2000). For this study, the variables were entered in the following order:
- Step 1. Counselor demographics—age, total years of experience, student/counselor ratio, and whether or not counselors were receiving any form of regular clinical supervision
- Step 2. Intrapersonal factors—TOC, EOC, andAOC
- Step 3. Organizational factors—FS, LDMA, CTPR, CPPR,RI,RC,andRA
Secondary analysis of the data was also completed. Frequency counts were conducted to check for possible data entry errors and residuals were evaluated to assess the data for extreme outliers. The results of all primary and secondary analyses are reported in the following section.
Next article with the result..
|
NEW ORLEANS link:http://www.oakl...
the link:http://www.oakleysungla...
Who is the link:http://www.topp...
Who is the link:http://www.topp...