This article was downloaded by: [Aston University]

On: 11 January 2014, At: 09:01

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Research in Childhood

Education
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/ujrc20

Teachers' Perceptions of Administrative

Support and Antecedents of Turnover
a a
Elizabeth Morgan Russell , Sue W. Williams & Cheryl Gleason-
a
Gomez
a
Texas State University , San Marcos, Texas
Published online: 05 Jul 2010.

To cite this article: Elizabeth Morgan Russell , Sue W. Williams & Cheryl Gleason-Gomez (2010)
Teachers' Perceptions of Administrative Support and Antecedents of Turnover, Journal of Research in
Childhood Education, 24:3, 195-208, DOI: 10.1080/02568543.2010.487397

To link to this article: http://dx.doi.org/10.1080/02568543.2010.487397

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Journal of Research in Childhood Education, 24: 195–208, 2010
Copyright © Association for Childhood Education International
ISSN: 0256-8543 print / 2150-2641 online
DOI: 10.1080/02568543.2010.487397

Teachers’ Perceptions of Administrative Support

and Antecedents of Turnover

Elizabeth Morgan Russell, Sue W. Williams, and Cheryl Gleason-Gomez

Texas State University, San Marcos, Texas
Downloaded by [Aston University] at 09:01 11 January 2014

The purpose of this pilot study was to determine the degree to which teachers’ age, perceptions of
fair pay, receipt of employer-sponsored health insurance, and administrative support, as operational-
ized by the Competing Values Framework, predicted antecedents of turnover. Teachers’ thoughts of
leaving their current job and commitment to a center were significantly predicted by perceptions of
their directors’ coordinating skills (e.g., dependable, skilled at enforcing rules and standards) and the
benefit of health insurance. Thoughts of leaving their current job also were predicted by fair pay. Job
tenure at a specific center was significantly predicted by perceptions of the director as an effective
coordinator and broker (e.g., skilled at seeking and gaining essential program resources), along with
the teacher’s age and perceptions of fair pay. The interpersonal skills quadrant of the competing val-
ues framework was not a significant predictor in any of the analyses. The value of the Competing
Values Framework for assessing teachers’ perceptions of administrative support is discussed.
Keywords: administrative support, child care, competing values framework, Early Childhood Work
Environment Survey, teacher turnover

Teacher turnover has been identified as a major factor undermining the quality of early care
and education (ECE) programs (DeVita, Twombly, & Montilla, 2002; Helburn, 1995; Howes,
Phillips, & Whitebook, 1992). High turnover undermines quality in a variety of ways, including
disruptions in teacher-child relationships (Howes, Hamilton, & Phillipsen, 1998) and in the roles
and relationships of teachers who remain at the center (Hamrick, 2000). In these ways, program
quality is negatively affected by turnover (Helburn, 1995; Phillips, Mekos, Scarr, McCartney,
& Abbott-Shin, 2000; Whitebook, Howes, & Phillips, 1990). Conversely, the quality of the ECE
program may affect whether teachers leave their jobs, thus creating a cyclical effect to the teacher
turnover process (Whitebook & Sakai, 2003). Regardless of the direction of the connections
between teacher turnover and program quality, it is generally agreed that turnover represents a
negative influence. And, most important, research indicates that program quality influences chil-
dren’s developmental outcomes (Burchinl, Peisner-Feinberg, Bryant, & Clifford, 2000; Helburn,
1995; National Institute of Child Health and Human Development, 2000).
Research has consistently identified wages and salaries as one of the most salient factors
impacting teachers’ job decisions related to turnover (Hale-Jinks, Knopf, & Kemple, 2006;

Submitted July 22, 2009; accepted February 10, 2010.

Address correspondence to Elizabeth Morgan Russell, Texas State University, Family and Consumer Sciences, 601
University Drive, San Marcos, TX 78666. E-mail: [email protected]
 196 RUSSELL, WILLIAMS, GLEASON-GOMEZ

Phillips et al., 2000; Whitebook & Bellm, 1999; Whitebook & Sakai, 2003). However, the
lack of employer-paid job benefits has inconsistently been identified as contributing to turnover
(Kontos & File, 1992; Whitebook & Sakai, 2003). Professional benefits include such things as
paid educational opportunities; personal benefits include payment for staff/family health insur-
ance or retirement pensions (Hale-Jinks et al., 2006). Employer payment of staff benefits is
limited in ECE programs in the United States (DeVita et al., 2002). In a 4-year longitudinal
study, Whitebook and Sakai (2003) reported that less than one half of teaching staff received
health benefits or a pension plan through work. However, they did not find paid health insurance
to be a significant factor in determining teachers’ job decisions.
Teacher age has been associated with turnover in previous studies (Manlove & Guzell, 1997;
Stremmel, Benson, & Powell, 1993). Older teachers are more likely to stay employed with the
same center than are younger teachers (Deery-Schmitt & Todd, 1995). Additionally, mature
Downloaded by [Aston University] at 09:01 11 January 2014

teachers view work and leadership differently than do younger teachers (Berl, 2005).
Another factor contributing to teacher turnover has been characterized as inadequate adminis-
trative support (Carter, 2000; Hale-Jinks et al., 2006; Whitebook & Sakai, 2003). Administrative
support has been loosely described in the ECE literature in a variety of ways, such as helpful
supervision (Kontos & File, 1992), collaboration with administrators (Carter & Curtis, 1998), and
creation of a sense of community (Bloom, 1995; Sciarra & Dorsey, 2003). However, with only
a very few exceptions (Bloom, 1996a; Quinn & Rorbaugh, 1981, 1983), the term administrative
support has not been operationally defined and is, therefore, open to individual interpretation.

TURNOVER PROCESS

A growing body of literature indicates that quitting a job is actually a process that occurs over a
period of time and that is influenced by external and internal factors (Manlove & Guzell, 1997).
Manlove and Guzell (1997) found that the availability of other jobs, years in child care, and
intention to leave were significant predictors of actual turnover in a sample of child care teachers.
Deery-Schmitt and Todd (1995) also identified job tenure as a predictor of turnover; those who
were on the job longer were more likely to remain on the job. Todd and Deery-Schmitt (1996)
identified the cumulative processes through which work environments might influence turnover.
They found that negative attitudes (often associated with teacher burnout) influenced teachers’
increasing cognitive withdrawal processes, which included thinking of quitting, searching for a
new job, and then actually quitting. Stremmel (1991) and Karsh, Booske, and Sainfort (2005)
examined the relationship of job commitment and job satisfaction to intention to leave. Although
both studies found job satisfaction and job commitment negatively related to intention to leave,
job commitment was the strongest predictor of turnover intentions.
In the current study, thoughts of leaving one’s current job, intention to leave, job commit-
ment, and job tenure were conceptualized as antecedents of turnover (Stremmel, 1991). The
current study focused on whether teachers’ perceptions of administrative support, as measured
and quantified by Quinn and Rorbaugh (1981, 1983), would significantly affect these antecedents
of turnover. Identifying predictors of the antecedents of turnover would provide directors with
information they could use to modify the work environment and make it more conducive to child
care staff remaining employed at their current center. Unlike improving wages and job benefits,
which may be beyond the control of directors, changing the level of support provided to teachers
should be within the capabilities of skilled administrators.
 ADMINISTRATIVE SUPPORT AND TURNOVER 197

MEASURES OF ADMINISTRATIVE SUPPORT

Two instruments that measure aspects of administrative support are the Early Childhood Work
Environment Survey (ECWES; Bloom, 1996a) and the competing values framework (CVF;
Quinn & Rorbaugh, 1983). The ECWES assesses staff’s perceptions of the organizational climate
of a work environment, including administrative support. The CVF was designed to measure cri-
teria associated with organizational effectiveness and has been used to assess staff’s perceptions
of and expectations for their supervisor’s administrative behaviors. Additional information about
each measure follows.

Early Childhood Work Environment Survey

Downloaded by [Aston University] at 09:01 11 January 2014

A broad characterization of administrative support is subsumed within the ECWES (Bloom,

1996a). The ECWES delineates factors related to a program’s organizational climate and
subsequent staff perceptions of these factors. For example, the ECWES measures the collec-
tive perceptions of staff on 10 dimensions of organizational climate (e.g., opportunities for
professional growth, supervisor support, reward system, etc.). The ECWES also compares
staff’s perceived and desired influence on decision-making, and their perceived and ideal work
environment.
Bloom (1966a) constructed organizational climate quite broadly; therefore, virtually all
aspects of the staff work environment are addressed. The 10 categories of the ECWES include
the following four examples: staff collegiality, supervisor support, physical environment, and
rewards system. One aspect of teacher behavior that Bloom studied relative to the ECWES was
teacher turnover (Bloom, 1996b). She found turnover related to the ECWES category, to teacher
commitment to the program, and to the total organizational climate score. The total alpha coef-
ficient for internal consistency of the 10 climate dimensions reported in the literature is .95. For
the current study, selected items of the ECWES were used as the dependent variables indicating
a teacher’s status regarding the antecedents of turnover.

Competing Values Framework

The CVF was originally developed to identify the criteria underlying organizational effec-
tiveness within the business management field (Quinn & Rorbaugh, 1983). The CVF was
subsequently used to study a wide range of organizational phenomena, including organiza-
tional culture (Howard, 1998), leadership, and decision-making (Hart & Quinn, 1993; Quinn &
Anderson, 1984). The CVF was constructed around two distinct dimensions representing com-
peting organizational values: internal/external and flexibility/control. These combinations result
in four quadrants of organizational activity (each with two subcategories) representing four dis-
tinct styles of organizational leadership and subsequent effectiveness (Edwards, 1987; Quinn &
Rorbaugh, 1981, 1983; Rojas, 2000). The CVF, although originally developed for use in business
management, is appropriate for use with various types of organizations (Lamond, 2003).
The four quadrants of CVF administrative behaviors are as follows: boundary spanning,
directing, coordinating, and human relations (Quinn, 1988). The boundary spanning quadrant
encompasses the innovator and broker roles and is focused on making effective organizational
 198 RUSSELL, WILLIAMS, GLEASON-GOMEZ

changes and acquiring essential resources (Edwards, Austin, & Altpeter, 1998). The human rela-
tions quadrant encompasses two roles (mentor and facilitator) that ensure the organization has
a competent workforce (e.g., effective communication patterns, staff are valued and motivated,
staff receive mentoring and training) (Edwards et al., 1998). The coordinating skills quadrant is
focused on maintaining the organization’s structure (e.g., budgeting and fiscal controls, schedul-
ing staff, quality control; Edwards et al., 1998). The directing skills quadrant is characterized
by activities (e.g., strategic planning and goal setting) that improve organizational efficiency and
effectiveness (Edwards et al., 1998). It is of interest to note that much of the applied literature
targeted at improving ECE directors’ administrative effectiveness is focused on competencies
similar to those subsumed within the human relations quadrant of the CVF.
The CVF is a 36-item, self-administered questionnaire that assesses managers’ administra-
tive behaviors. In the current study, it was used to assess teachers’ perceptions of their directors’
Downloaded by [Aston University] at 09:01 11 January 2014

administrative support. Participants responded to each item using a 7-point Likert-type scale.
Selected items were minimally modified in terminology to be more applicable to the child care
setting (e.g., the term program was used in lieu of unit). The CVF has eight subscales, each
assessing a possible management role (innovator, broker, producer, director, coordinator, mon-
itor, facilitator, and mentor), that cluster into four quadrants (Edwards et al., 1998). Quinn and
Spreitzer (as cited in Rojas, 2000) confirmed the validity of these four quadrants by testing them
on 796 executives, “where the analysis was performed by comparing two types of competing
value scales” (Rojas, 2000, p. 100). Kalliath, Bluedorn, and Gillespie (1999) used structural
equation modeling on a sample of 300 managers and supervisors to validate the CVF model. See
Table 1 for the coefficient alphas for each subscale in the current study. Mean replacement was
used for missing data.
The purpose of this pilot study was to determine the degree to which age, perceptions of fair
pay, employer-sponsored health insurance, and administrative support, as conceptualized by the
four quadrants of the CVF, helped predict the identified antecedents of turnover as delineated
by the ECWES. The antecedents of turnover were thoughts of leaving their current ECE center,
feelings of commitment to their current ECE center, intention to work at least 2 more years at the
same ECE center, and job tenure at a particular center.

TABLE 1
Coefficient Alphas for Each Subscale (N = 78)

Variable Alpha

Boundary spanning quadrant

Innovatora .91
Brokerb .87
Human relations quadrant
Mentor .92
Facilitatorc .86
Coordinating skills quadrant
Coordinatord .90
Producere .78
Directing skills quadrant
Director .92
Monitorf .76

Note. a. n = 77; b. n = 72. c. n = 75. d. n = 77. e. n = 76. f. n = 74.

 ADMINISTRATIVE SUPPORT AND TURNOVER 199

METHOD

Participants

The participants were recruited from two adjacent counties in central Texas. Seventy-eight teach-
ers who were employed full-time at a licensed child care center were included in this pilot study.
The teachers were recruited from child care centers randomly selected, using random numbers,
from a list of centers in good standing with the Texas Department of Family and Protective
Services (TDFPS); these centers had no findings of noncompliance from TDFPS. Only 6.9% of
Texas child care centers are accredited by the National Association for the Education of Young
Children (NACCRRA; 2009), therefore, it would have been very challenging to recruit a sample
of teachers from accredited programs for this pilot study. Because at least seven staff members
Downloaded by [Aston University] at 09:01 11 January 2014

from each ECE center were required to respond to the ECWES for the measure to be scored,
only programs licensed to serve 100 or more children were included in the selection pool. At
the 90 centers, 24 directors, all female, agreed to allow their teachers to be recruited for the
study—a 27% response rate. Most directors were contacted 3 to 5 times in an effort to increase
the response rate.

Measures

The four dependent variables were selected from the ECWES and are listed here with their exact
wording as well as the dependent variable name in parentheses: “I intend to work here at least
two more years” (stay), “I often think of quitting” (leaving job), “I feel very committed to this
center” (commit), and “Months worked for present employer” (job tenure). The first three of the
aforementioned dependent variables were dichotomous (yes/no) and the fourth was continuous.
There were 11 independent variables, three of which, age, fair pay (yes/no), and health insur-
ance (do have/don’t have), were selected from the ECWES. Fair pay (i.e., “Salaries . . . are
fair, considering the center’s income”) was chosen as an independent variable, because recent
research identified compensation as a significant predictor of turnover (Whitebook & Sakai,
2003). Although health benefits were not found to be consistently predictive of turnover in
previous research (e.g., Whitebook & Sakai, 2003), having health insurance was included as
an independent variable in this pilot study because of the salience of health care and health
insurance in the broader culture at the time the study was conducted in 2007 to 2008. Age was
included as an independent variable, because this sample of teachers was mature (i.e., older),
and as previously stated, older teachers are more likely to stay employed with the same cen-
ter than are younger teachers. The remaining independent variables were the eight continuous
subscales of the CVF: Facilitator, Innovator, Producer, Coordinator, Mentor, Broker, Director,
and Monitor, because they represent the different aspects of directors’ total administrative
support.

Procedure

The purpose of the current study was explained via phone to directors of randomly selected
ECE centers, with permission obtained from 24 directors prior to approaching the teaching staff.
A graduate student, or one of the study authors, traveled to each center at a prearranged time
 200 RUSSELL, WILLIAMS, GLEASON-GOMEZ

to discuss the study purpose and procedures in a group setting. The measures, which included
the consent form, a demographic form, the ECWES, and the CVF, also were introduced, and
instructions for completing the forms were provided. Although the director arranged the group
meeting, she was not present during the discussion of the study. To further protect the anonymity
of the teachers who chose to participate through completing all forms, a sealed box with an
open slit on top was left at each center. All of the teachers, whether they chose to participate
or not, were instructed to place the measures in an envelope provided for that purpose, seal
the envelope, and place it in the box. Therefore, the director could not determine which teach-
ers actually participated in the study. After 2 weeks, the box was collected, emptied, resealed,
and returned to the center for an additional 2 weeks. The box was removed at the end of the
second 2-week period. Seventy-eight teachers (excluding teacher aides) completed the ECWES
and CVF.
Downloaded by [Aston University] at 09:01 11 January 2014

RESULTS

Participants

Participants had a mean age of 35 years (SD = 11.36) and had worked in the field of early
childhood for a mean of 109 months (SD = 81.65). The ethnic breakdown of the 78 participants
was as follows: African American 3.8% (n = 3), Euro American 60% (n = 47), Asian American
2.6% (n = 2), Hispanic American 21.8% (n = 17), and other ethnicity 1.3% (n = 1). Eight
teachers (10.3%) did not provide information about their ethnicity. Twenty participants (25.6%)
had obtained a bachelor’s degree, 18 (23.1%) had a high school diploma or equivalent, 6 (7.7%)
had earned an associate’s degree, and an additional 34 (43.6%) had achieved some level of college
education.

Teacher Perceptions of Directors’ Administrative Support

Analyses indicated that overall, the participants evaluated their directors as being relatively
skilled at all eight CVF administrative support behaviors. The means on the eight subscales
ranged from 4.39 to 5.32. The means and standard deviations are found in Table 2.

Predictors of Stay

Forty six (59%) of 78 participants reported they intended to stay at the same center for at least
2 more years. Analyses were conducted to identify significant bivariate relationships; none was
found (see Table 3). All 11 independent variables were included in binary logistic regression
for Stay. Regression results indicate that the overall model of the 11 predictor variables was
statistically nonsignificant in distinguishing between participants who did and did not agree that
they planned to stay at the center for two more years. Due to space constraints, the results are not
presented here. (Results are available from the first author.)
 ADMINISTRATIVE SUPPORT AND TURNOVER 201

TABLE 2
Means and Standard Deviations of Continuous Dependent
and Independent Variables (N = 78)

Variable M (SD)

Dependent
Months workeda,b 39.22 (52.88)
Independent
Age 34.93 (11.37)
Fair payc,d 2.92 (1.48)
Facilitator 4.59 (1.53)
Innovator 4.90 (1.59)
Producer 5.25 (1.33)
Coordinator 5.13 (1.44)
Downloaded by [Aston University] at 09:01 11 January 2014

Mentor 5.27 (1.64)

Broker 4.39 (1.87)
Director 5.32 (1.45)
Monitor 5.00 (1.44)

Note. a. Months worked = Months worked for current employer.

b. n = 75.
c. Fair pay = Perceptions of fairness of pay.
d. n = 77.

TABLE 3
Bivariate Relationships Between Dichotomous DVs and Independent Variables (N=76)

Dependent Variable

Independent Variables Stay Quit Commit

Health Insurancea χ2 2.05 8.08∗∗ 12.93∗∗∗

t test
Ageb 1.67 −1.05 1.47
Fair Payc 1.02 −3.79∗∗∗ 1.51
Facilitator 1.21 −3.80∗∗∗ 3.66∗∗∗
Innovator 1.24 −3.78∗∗∗ 3.73∗∗∗
Producer 0.39 −2.93∗∗ 3.40∗∗
Coordinator 1.14 −5.67∗∗∗ 4.70∗∗∗
Mentor 1.65 −3.55∗∗∗ 3.27∗∗
Broker 0.72 −2.57∗ 2.59∗
Director 1.16 −4.25∗∗∗ 3.68∗∗∗
Monitor −0.05 −3.41∗∗ 2.46∗

Note. a n = 72. b n = 73. c n = 75.

∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001.
 202 RUSSELL, WILLIAMS, GLEASON-GOMEZ

Predictors of Commit

Fifty-two (67%) of 78 participants reported feeling very committed to the ECE center at which
they were currently employed. Analyses were conducted to identify significant bivariate rela-
tionships; all of the independent variables, with the exception of age, were significantly related
to commit. The results also are found in Table 3. All 11 independent variables were included in
the logistic regression for commit. Regression results indicate that the overall model of the 11
predictor variables was statistically significant in distinguishing between participants who did
and did not agree that they felt committed to the center (–2 log likelihood = 56.67). The model
correctly classified 80% of the cases. Regression coefficients are presented in Table 4. Wald
statistics indicated that health insurance and coordinator significantly predicted commit. Odds
ratios for health insurance indicated that the participants were 12 times more likely to report
Downloaded by [Aston University] at 09:01 11 January 2014

commitment to their ECE center if they had health insurance. Odds ratios indicate that for every
level increase in the coordinator variable, the participants were 365% more likely to report that
they “were committed to the center.”

Predictors of Leaving Job

Twenty (26%) of the 78 participants reported they often thought of leaving their current job.
Analyses were conducted to identify significant bivariate relationships for inclusion in regression
analysis of leaving job: Age and fair pay were not significantly related to leaving job, whereas
the remaining nine variables were. The results also are found in Table 3. All 11 independent vari-
ables were included in the logistic regression for leaving job. Regression results indicate that the
overall model of the 11 predictor variables was statistically significant in distinguishing between
participants who did and did not agree they often thought of leaving their current job (–2 log
likelihood = 37.23). The model correctly classified 86% of the cases. Regression coefficients
are presented in Table 5. Wald statistics indicated that health insurance, fair pay, and coordinator
significantly predicted leaving job. Odds ratio indicate that participants without health insur-
ance were 92% more likely to report they often thought of leaving their current job. Those who
reported the pay was unfair were 60% more likely to report having often thought of leaving their
current job. Odds ratios indicate that for every level decline in the coordinator variable, the par-
ticipants were 87% more likely to report that they often thought of leaving their current job.

Predictors of Job Tenure

A linear multiple regression analysis was conducted to determine if the 11 independent variables
influenced the number of months the participants reported working for their current employer.
Prior to conducting the regression, the data were screened for multivariate outliers, multi-
collinearity, and singularity. One outlier was excluded from analysis. Initial correlations revealed
low to high intercorrelations among the variables ranging between –.01 and .88, indicating the
possible presence of multicollinearity among the eight subscales of the CVF. Table 6 presents the
intercorrelation matrix. However, the absence of multicollinearity among the remaining variables
was confirmed by tolerance statistics ranging from .15 to .92 and by variance inflation factors
 ADMINISTRATIVE SUPPORT AND TURNOVER 203

TABLE 4
Summary of Logistic Regression Analysis for Variables Predicting Commit (N = 69)

Variable b SE b Wald df Exp (b)

Age 0.04 0.03 1.22 1 1.04

Health Insurance 2.52 0.84 9.03 1 12.42∗∗
Fair Pay 0.19 0.29 0.40 1 1.20
Facilitator −0.42 0.51 0.69 1 0.66
Innovator 0.15 0.45 0.11 1 1.56
Producer 0.42 0.55 0.59 1 1.53
Coordinator 1.29 0.62 4.41 1 3.65∗
Mentor −0.02 0.37 0.00 1 0.98
Broker 0.48 0.28 2.93 1 1.61t
Director −0.21 0.58 0.13 1 0.81
Downloaded by [Aston University] at 09:01 11 January 2014

Monitor −0.77 0.51 2.31 1 0.46

t p < .10, ∗ p < .05, ∗∗ p < .01.

TABLE 5
Summary of Logistic Regression Analysis for Variables Predicting Leaving Job (N = 69)

Variable b SE b Wald df Exp (b)

Age −0.06 0.05 1.71 1 0.94

Health insurance −2.48 1.15 4.65 1 0.08∗
Fair pay −0.93 0.40 5.34 1 0.40∗
Facilitator −0.10 0.66 0.02 1 0.91
Innovator 0.33 0.69 0.24 1 1.40
Producer 0.62 0.75 0.69 1 1.87
Coordinator −2.04 0.93 4.79 1 0.13∗
Mentor 0.08 0.50 0.26 1 1.08
Broker −0.23 0.40 0.31 1 0.80
Director 0.55 0.88 0.39 1 1.74
Monitor 0.19 0.64 0.09 1 1.20
∗ p < .05.

TABLE 6
Intercorrelations Among Variables (N = 68)

Variable 1 2 3 4 5 6 7 8 9 10 11 12

Job tenure — .34∗∗ .22∗ –.19 .10 .04 –.07 .12 –.02 .23∗ .05 –.03
Age — .08 .05 .94 .11 –.02 .04 .05 .13 .06 –.02
Health Insurance — .05 .18 .18 .12 .21∗ .09 –.01 .28∗ .11
Fair pay — .38∗∗∗ .34∗∗ .27∗ .39∗∗∗ .27∗ .18 .36∗∗∗ .32∗∗
Facilitator — .81∗∗∗ .79∗∗∗ .85∗∗∗ .75∗∗∗ .70∗∗∗ .77∗∗∗ .71∗∗∗
Innovator — .76∗∗∗ .81∗∗∗ .73∗∗∗ .57∗∗∗ .87∗∗∗ .74∗∗∗
Producer — .76∗∗∗ .69∗∗∗ .58∗∗∗ .81∗∗∗ .79∗∗∗
Coordinator — .75∗∗∗ .54∗∗∗ .88∗∗∗ .74∗∗∗
Mentor — .54∗∗∗ .71∗∗∗ .62∗∗∗
Broker — .54∗∗∗ .56∗∗∗
Director — .80∗∗∗
Monitor —
∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001.
 204 RUSSELL, WILLIAMS, GLEASON-GOMEZ

TABLE 7
Summary of Regression Analysis for Variables Predicting Job Tenure (N=68)

Variable b SE b β t test

Age 1.33 0.52 .28 2.56∗

Health Insurance 24.02 13.99 .20 1.72t
Fair Pay −9.89 4.26 −.27 −2.32∗
Facilitator −3.41 9.43 −.10 −0.36
Innovator −0.31 8.80 −.01 −0.35
Producer −14.25 9.16 −.35 −1.56
Coordinator 23.75 10.96 .61 2.17∗
Mentor −7.42 6.04 −.21 −1.23
Broker 11.34 4.57 .39 2.48∗
Director −3.52 11.70 −.09 −0.31
Downloaded by [Aston University] at 09:01 11 January 2014

Monitor −2.52 7.10 −.07 −0.35

Note. R2 = .38.
t p < 10. ∗ p < .05.

ranging from 1.09 to 8.56. According to Mertler and Vannatta (2005), tolerance statistics above
.10 and variance inflation factors (VIF) values below 10 indicate that multicollinearity does not
pose a problem. The model, Model 1, R2 = .38, R2 adj = .25, F(11, 56) = 3.05, p < .01, accounted
for about 38% of the variance in job tenure. Statistically significant amounts of variance were
accounted for by four independent variables: age, fair pay, coordinator, and broker. A summary
of the regression model is presented in Table 7. The older the participants, the longer they had
been employed at the center. The longer the participants had been employed at the center, the
less likely they were to perceive the pay as fair. The longer the participants had been employed
at the center, the higher the coordinator and broker scores they reported for their director.

DISCUSSION

Low salaries/wages (Helburn, 1995; Whitebook et al., 1990; Whitebook & Sakai, 2003) and
inadequate benefits (Kontos & File, 1992) predict teacher turnover, although the significance
of benefits has not been consistently found (Whitebook & Sakai, 2003). Therefore, it was not
surprising that pay perceived as unfair and a lack of employer-sponsored health benefits pre-
dicted thoughts of leaving their current job among participants in the current study. However,
the results of the current study go beyond the findings of previous turnover-related research by
examining how teachers’ views of their director’s administrative support affect employment deci-
sions. Teachers who perceived their director to be less skilled as a coordinator, that is, those who
rated their director as less dependable, consistent, and skilled at scheduling and enforcing rules
and standards, also reported they often thought of leaving their current job. Conversely, job com-
mitment and job tenure were predicted by perceptions of the director as supportive through her
skills as a coordinator. Teaching young children is stressful (Deery-Schmitt & Todd, 1995), and a
skilled administrator can make it less stressful, increase motivation, and, thus, facilitate longevity
and commitment (Todd & Deery-Schmitt, 1996).
 ADMINISTRATIVE SUPPORT AND TURNOVER 205

Teacher job commitment to an ECE program also was facilitated by equitable pay and
employer-sponsored health insurance. Teachers who reported they had health insurance were
12 times more likely to report feeling committed to their current position. Of course, percep-
tions of pay as unfair predicted teachers’ thoughts of leaving their current job. The finding that
health insurance predicted participants’ job commitment was undoubtedly influenced by a gen-
eral awareness in the United States of the large numbers of individuals lacking this job benefit,
and about the growing cost of health care. Additionally, a director who obtains health insurance
for her teachers may create goodwill toward herself, which enhances motivation and increases
job commitment (Catapono, 2001).
Job tenure was predicted by teachers’ perceptions of the director as an effective broker (i.e.,
competent in seeking and gaining program resources that teachers perceived as supportive of their
role) (Edwards et al., 1998). The longer that teachers had been employed with an ECE program,
Downloaded by [Aston University] at 09:01 11 January 2014

the higher the score assigned to the director’s ability as a broker. Intriguingly, the longer that
participants worked at a center, the less likely they were to perceive their pay as fair. Although
teachers with a longer tenure perceived their pay as unsatisfactory, perceptions of the director as
an effective broker apparently helps to offset teachers’ dissatisfaction with wages.
This group of participants was mature and had worked an average of 3.27 years at their jobs.
Previous research (Manlove, 1993; Manlove & Guzell, 1997; Petty, Brewer, & Brown, 2005;
Stremmel et al., 1993) has found age to be an inconsistent predictor of turnover. In this sample,
age was only predictive of years worked at a center. However, the age range of the teachers
indicates that directors should strive to address the needs and interests of young and mature
workers in different ways (Berl, 2005).
The human relations quadrant (i.e., mentor and facilitator administrative arenas) was not
shown to be significant in any of the analyses. This quadrant is conspicuously absent because
it encompasses those traits and competencies that are often associated with the ideal ECE pro-
gram administrator (i.e., one who is empathic and caring, who possesses good interpersonal
and communication skills, and who works to build staff cohesion and foster cooperation). This
quadrant is generally subsumed under the broad category of administrative support in the ECE
literature, and so its absence has been related to staff turnover in other research (Carter & Curtis,
1998; Kontos & File, 1992; Sciarra & Dorsey, 2003).

Implications and Limitations of the Study

Teacher turnover has deleterious effects on the quality of care provided to children (Helburn,
1995; Whitebook et al., 1990; Whitebook & Sakai, 2003), and it is understood that poor-quality
care can negatively affect the development of young children (Phillips et al., 2000; Whitebook
et al., 1990). As found in previous studies, pay and benefits were linked to two antecedents of
turnover: participants’ thoughts of leaving their current job and feelings of commitment to their
current teaching position. Much has been written about the importance of adequate compensation
to teacher stability (e.g., Whitebook et al., 1990; Whitebook & Sakai, 2003) and will not be
repeated here. Rather, the implications of the findings related to the CVF is discussed.
Teacher longevity has been linked to perceived administrative support (Catapono, 2001;
Kontos & File, 1992). However, operationalizing administrative support in the child care arena
has been an elusive pursuit. According to the results of the current study, the CVF shows promise
 206 RUSSELL, WILLIAMS, GLEASON-GOMEZ

as a measure for assessing teachers’ perceptions of their director’s competencies in eight adminis-
trative arenas: facilitating, innovating, producing, coordinating, mentoring, brokering, directing,
and monitoring. In turn, two of these arenas, coordinating and brokering, were predictive of
teachers’ thoughts of leaving their current job, feelings of commitment, and years of service.
Directors committed to improving program quality by reducing turnover may be well served to
take advantage of training opportunities for enhancing their skills in the aspects of administrative
support comprising the CVF.
However, additional work is needed before these findings can be generalized beyond this
pilot study, due to limitations concerning the sample’s size and nature. The current study should
be replicated with a larger and more diverse sample. Two obstacles had to be overcome to
obtain these data. First, the director and/or other corporate entity had to give permission for
us to attempt to recruit teachers within a center, and 27% did so. Second, teachers within these
Downloaded by [Aston University] at 09:01 11 January 2014

24 programs had to decide whether to participate.

As previously stated, these participants were mature, well-educated, and experienced. It may
well be that the teachers who decided to participate were better educated than those who chose
not to participate; however, we were unable to gather any data on teachers who decided not to
participate. It is also unknown whether the well-educated teachers were certified by the Texas
Education Administration, as these data were not obtained.
Additionally, participants perceived their directors as being relatively skilled in all eight
administrative roles, indicating overall satisfaction with administrative support. However, due
to study methodology, this was a self-selected sample. The 24 directors who agreed to partic-
ipate may have perceived their teaching staff as being satisfied with the effectiveness of their
administrative support. Also, based upon oral feedback given to the researchers by teachers
declining to participate in the study, those who might have rated their directors as less skilled
in the administrative support roles were possibly reluctant to participate because of concerns
regarding confidentiality. These concerns remained despite all efforts made to ensure and assure
the study participants of their anonymity.
One other limitation had to do with the surveys. Because items were selected out of the orig-
inal instruments, with some modified for the purposes of the current study, the reliability and
validity of the results may be affected. Nonetheless, the potential value of the CVF to the field of
ECE merits further exploration.

REFERENCES

Berl, P. S. (2005). Mature teachers matter. Child Care Information Exchange, 165, 10–13.
Bloom, P. J. (1995, January). Building a sense of community: A broader view. Child Care Information Exchange, 101,
47–50.
Bloom, P. J. (1996a). Improving the quality of work life in early childhood settings: Resource guide and technical manual
for the early childhood work environment survey (Rev. ed.). Wheeling, IL: National-Louis University.
Bloom, P. J. (1996b). The quality of work life in NAEYC accredited and nonaccredited early childhood programs. Early
Education and Development, 7(4), 301–317.
Burchinal, M. R., Peisner-Feinberg, E., Bryant, D. M., & Clifford, R. (2000). Children’s social and cognitive devel-
opment and child-care quality: Testing for differential associations related to poverty, gender, or ethnicity. Applied
Developmental Science, 4(3), 149–165.
Carter, M. (2000). What do teachers need most from their directors? Child Care Information Exchange, 136, 98–101.
 ADMINISTRATIVE SUPPORT AND TURNOVER 207

Carter, M., & Curtis, D. (1998). The visionary director: A handbook for dreaming, organizing, and improvising in your
center. St. Paul, MN: Redleaf Press.
Catapono, S. (2001). Why do teachers stay? Child Care Information Exchange, 142, 66–69.
Deery-Schmitt, D. M., & Todd, C. M. (1995). A conceptual model for studying turnover rate among family child care
providers. Early Childhood Research Quarterly, 10, 121–143.
DeVita, C., Twombly, E., & Montilla, M. (2002). Toward better child care worker compensation: Advocacy in three
states. (ERIC Documentation Reproduction Service No. ED 464715)
Edwards, R. L. (1987). The competing values approach as an integrating framework for the management curriculum.
Administration in Social Work, 11, 1–13.
Edwards, R. L., Austin, D. M., & Altpeter, M. A. (1998). Managing effectively in an environment of competing values.
In R. L. Edwards, D. M. Austin, & M. A. Altpeter (Eds.), Skills for effective management of nonprofit organizations
(pp. 5–21). Washington, DC: National Association of Social Workers Press.
Hale-Jinks, C., Knopf, H., & Kemple, K. (2006). Tackling teacher turnover in child care. Childhood Education, 82,
219–226.
Downloaded by [Aston University] at 09:01 11 January 2014

Hamrick, J. (2000). Reduce staff turnover through effective interviewing. Child Care Information Exchange, 147, 26–28.
Hart, S. L., & Quinn, R. E. (1993). Roles executives play: CEOs, behavioral complexity, and firm performance. Human
Relations, 46, 543–574.
Helburn, S. W. (1995). Cost, quality, and child outcomes in child care centers: Technical report. Denver: University of
Colorado, Center for Research in Economic and Social Policy, Department of Economics.
Howard, L. W. (1998). Validating the competing values model as a representation of organizational cultures. International
Journal of Organizational Analysis, 6, 231–250.
Howes, C., Hamilton, C. E., & Phillipsen, L. C. (1998). Stability and continuity of child-caregiver and child-peer
relationships. Child Development, 69(2), 418–426.
Howes, C., Phillips, D. A., & Whitebook, M. (1992). Thresholds of quality: Implications for the social development of
children in center-based child care. Child Development, 63, 449–460.
Kalliath, T. J., Bluedorn, A. C., & Gillespie, D. F. (1999). A confirmatory factor analysis of the Competing Values
Instrument. Educational and Psychological Measurement, 59(1), 143–158.
Karsh, B., Booske, B. C., & Sainfort, F. (2005). Job and organizational determinants of nursing home employee
commitment, job satisfaction, and intent to leave. Ergonomics, 48, 1260–1281.
Kontos, S., & File, N. (1992). Condition of employment, job satisfaction, and job commitment among early intervention
personnel. Journal of Early Intervention, 16, 155–165.
Lamond, D. (2003). The value of Quinn’s competing values model in an Australian context. Journal of Managerial
Psychology, 18, 46–59.
Manlove, E. E. (1993). Multiple correlates of burnout in child care workers. Early Childhood Research Quarterly, 8(4),
499–518.
Manlove, E. E., & Guzell, J. R. (1997). Intention to leave, anticipated reasons for leaving, and 12-month turnover of child
care center staff. Early Childhood Research Quarterly, 12, 145–167.
Mertler, C. A., & Vannatta, R. A. (2005). Advanced and multivariate statistical methods (3rd ed.). Glendale, CA: Pyrczak
Publishing.
National Association of Child Care Resource and Referral Agencies. (2009). 2009 child care in the state of Texas.
Retrieved December 17, 2009, from www.naccrra.org/randd/data/docs/TX.pdf
National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network. (2000). The
relation of child care to cognitive and language development. Child Development, 71, 960–980.
Petty, G. C., Brewer, E. W., & Brown, B. (2005). Job satisfaction among employees of a youth development organization.
Child and Youth Care Forum, 34, 57–73.
Phillips, D., Mekos, D., Scarr, S., McCartney, K., & Abbott-Shin, M. (2000). Within and beyond the classroom door:
Assessing quality in child care centers. Early Childhood Research Quarterly, 15, 475–496.
Quinn, R. E. (1988). Beyond rational management.San Francisco: Jossey-Bass.
Quinn, R. E., & Anderson, D. F. (1984). Formalization as crisis: Transition planning for a young organization. In
J. R. Kimberly & R. E. Quinn (Eds.), New futures: The challenge of managing corporate transitions (pp. 11–28).
Homewood, IL: Dow Jones-Irwin.
Quinn, R. E., & Rohrbaugh, J. (1981). A competing values approach to organizational effectiveness. Public Productivity
Review, 5, 122–140.
 208 RUSSELL, WILLIAMS, GLEASON-GOMEZ

Quinn, R. E., & Rohrbaugh, J. (1983). A spatial mode of effectiveness criteria: Towards a competing values approach to
organizational analysis. Management Science, 29, 363–377.
Rojas, R. R. (2000). A review of models for measuring organizational effectiveness among for-profit and nonprofit
organizations. Nonprofit Management & Leadership, 11(1), 97–104.
Sciarra, D. J., & Dorsey, A. G. (2003). Developing and administering a child care center (5th ed.). Clifton Park, NY:
Thomson.
Stremmel, A. J. (1991). Predictors of intention to leave child care work. Early Childhood Research Quarterly, 6, 285–298.
Stremmel, A. J., Benson, M. J., & Powell, D. R. (1993). Communication, satisfaction, and emotional exhaustion
among child care center staff: Directors, teachers, and assistant teachers. Early Childhood Research Quarterly, 8(2),
221–233.
Todd, C. M., & Deery-Schmitt, D. M. (1996). Factors affecting turnover among family child care providers: A
longitudinal study. Early Childhood Research Quarterly, 11, 351–376.
Whitebook, M., & Bellm, D. (1999). Taking on turnover: An action guide for child care center teachers and
directors.Washington, DC: Center for the Child Care Workforce.
Downloaded by [Aston University] at 09:01 11 January 2014

Whitebook, M., Howes, C., & Phillips, D. (1990). The National Child Care Staffing Study. Final report: Who cares?
Child care teachers and the quality of care in America. Washington, DC: Center for the Child Care Workforce.
Whitebook, M., & Sakai, L. (2003). Turnover begets turnover: An examination of job and occupational instability among
child care center staff. Early Childhood Research Quarterly, 18, 273–293.