Inner City Teachers’ Sense of Efficacy
Towards Minority Students
Deborah M. Oh, Ph.D.
Assistant Professor
King Hall A2035
323-343-6155
Joshua M. Kim, M.A.
Doctoral Student
King Hall A2035
323-343-6155
Cynthia H. Leyva, B.A.
(213) 622-0669
Inner City Teachers’ Sense of Efficacy
Towards Minority Students
Annotation – The purpose of this study
was to explore the inner city teachers’ sense of efficacy and their perceptions
and attitudes towards Latino, language minority, and low socioeconomic status
students in three urban schools in
Abstract – Eighty-seven inner city
teachers in the
Inner City Teachers’ Sense of Efficacy Towards Minority Students
The extent to which teachers believe they are able to affect student performance is known as teacher efficacy (Gibson & Dembo, 1984). Efficacy is related to setting goals and to exercising effort and persistence in their attainment (Bandura, 1997). It has been linked to teachers’ instructional practices and attitudes toward students (Bender, Vail, & Scott, 1995; Gibson & Dembo, 1984; Midgley, Anderman & Hicks, 1995; Midgley, Feldlaufer, & Eccles, 1989).
Current reform issues in education place great emphasis on raising the achievement level of all students, particularly the low performance of minority students. Research on the efficacy of teachers report significant relationships between teachers’ sense of efficacy and higher student achievement and test scores. Teachers with high levels of efficacy are more likely to expect that all students can learn, and to feel responsible for that learning, than are teachers with low efficacy levels (Ashton & Webb, 1986, Tracz & Gibson, 1986).
While research has demonstrated that teacher expectations can affect students’ achievement and attitudes (Bamburg, 1994; Cotton & Wikelund, 1997; Hurley, 1997; Lumsden, 1997; Reeves & Taylor, 1993), Bamburg (1994) showed that some teachers do not expect much from the students in their schools. According to Tauber (1998) the effects of verbal and nonverbal expectations can be detrimental. Early childhood and elementary age children are most at risk in believing teachers consistently delivered evaluations as either positive or negative.
Other factors have been identified that differentiate opportunities for students, in turn influencing their academic achievement. Good and Brophy (1972) observed that teachers interact differently with low and high achievers in the classroom. Low achievers received general, sometimes insincere praise, less feedback, more interruptions, less attention, more criticism, less time to answer questions, were called on less often, smiled at less, given fewer nonverbal indicators of support, (such as head nodding), and were seated farther away. Alvidrez & Weinstein’ (1999) suggestion that the higher the socioeconomic status, the higher the teacher rates a child’s intelligence indicates that teachers’ preconceptions and beliefs can have an adverse effect on a child’s level of achievement. These misconceptions may be conscious or unconscious and reflect societal biases.
Teachers in urban schools increasingly find themselves guiding students who are very different from themselves. Cotton and Wikeland (1997) assert that educators differentiate among students based on gender, race/ethnicity, socioeconomic status, language, tracking, and negative comments about students. Teachers who exhibit such differential attitudes may not even be aware of their perceptions, yet some children will continue to be at a disadvantage in the classroom. Ethnic minority and poor children have been documented to be largely over-represented as targets of low expectations (Weinstein, 1995). According to Parson’s study (as cited in Caruthers, 1997) teachers praise and encourage white students, respond to them, and pay more attention to them than to Mexican-American students.
Teachers, in general, expect more from middle and upper class students than from working and lower-class backgrounds (Caruthers, 1997). It is easy to assess the socioeconomic status of students in schools where a majority of the students are enrolled in free and reduced lunch programs. Poverty becomes one more factor, along with race or ethnicity, to be associated with a low-status background. This can influence teacher expectations of children who belong to groups long impeded by legal and social discrimination, as well as historically poor records of academic achievement. In the classroom, as in the streets outside the schoolyard gate, we see their faces, or we strive to perceive more, in hearts and minds.
The purpose of
this study was to explore the sense of efficacy of inner city teachers and
their perceptions and attitudes towards Latino, language minority, and low
socioeconomic status students in three urban schools in
The questions for the study were:
Method
Participants
A pool of 87
classroom teachers from grades Pre-Kindergarten through 12 from three
inner-city schools in the
The demographics of the 87 teachers surveyed were: 27% male and 73% female, teaching in primary (preK-5, 26%), intermediate (junior high/middle, 39%) and upper grades (high school, 34%). Over half of the teachers were Latino (52.9%, followed by 25.3% White, 10.3% Asian, and 5.7% African American). Over sixty-five percent (65.5%) received a bachelor’s degree and 30% had received a master’s degree. About half of the sample specialized in either multiple subjects (17.2%) or BCLAD (19.5%), and the others specialized or had a credential in the following other areas: national certificate, special education, educational credential, and preschool permit. The average age of the teachers was 35, with teaching experience ranging up to 35 years (mean = 8.63, SD = 7.95 years). The average number of years at the present school was just over 4 years (range: less than 1 year to 18 years).
Instruments
and Procedure
Data were collected through a survey composed of three sections (see Appendix A). Participation was voluntary. In the first section, participants provided demographic information including age, gender, ethnicity, highest level of education, teaching level, and years of teaching experience. Teachers were instructed not to use names and each data sheet was numbered to ensure confidentiality. The second section consisted of a Teacher Efficacy Scale composed of 23 question items. The participating teachers were asked to indicate their degree of agreement with each item by responding on a Likert scale from 1 strongly disagree to 5 strongly agree. The original scale used in the study was created by Gibson & Dembo (1984, later modified by Deemer & Minke in 1999). Deemer & Minke’s modification was based on their finding that the efficacy scale influenced responses by the wording of the question items (either positive or negative). Therefore, both positive and negative items were used in the questionnaire to off-set possible differences as a result of wording in the current study.
The final section of the survey was designed to gather participants’ explanations for their efficacy, beliefs and the factors they perceived to influence these beliefs and perceptions. The first item was designed to represent the main idea of the internal scale: “How confident are you that you can change student learning in positive ways?” Participants responded using a scale ranging from a 5-point Likert scale: 1 = not at all confident to 5 = very confident. This item was followed by the open-ended question, “Why did you rate your level of confidence as you did?” The second scaled item parallels the external dimension: “To what extent do/will factors beyond your control influence your confidence to impact student learning?” It was answered using a similar 5-point Likert scale: 1 = not at all and 5 =very much. This item was followed by another open-ended question: “What are these factors?”
Results
There were two emerging themes from the 23 efficacy item questions on the survey: instructional practices and expectations. Through principle components analysis, these themes were constructed from five and six question items, respectively, to be used for analysis (see Table 1). Questions on instructional practices included items such as: “If a student in my class becomes disruptive and noisy, I feel assured that I know the technique to redirect him quickly”; “When the grades of my students improve, it is usually because I found more effective teaching approaches.” Questions on expectations were questions on teachers’ expectation of students’ potential for learning such as: “Nearly all my students will be at or above grade level by the end of this year” and their expectation on a specific task on the national tests such as: “Most students in my school will perform at about the national average in academic achievement.” Each construct, instructional practices and expectations, showed Cronbach’s alphas: .65 and .72, respectively.
The survey items Confidence level and external factors were measured on a 5-point scale for the following questions, respectively: “How confident are you that you can change student learning in positive ways?” and “To what extent do/will factors beyond their control influence their confidence to impact student learning?”
Instructional practice, expectations, confidence level, external factors, and the total average of all 23 items, referred to as efficacy in the study, were used for descriptive analyses and means comparison by teachers’ background measures. Efficacy and external factors measures were used as outcome variables for predictability in the regression analyses.
Descriptive
Analysis
Expectations, instructional practices and
efficacy level. Although teachers were not sure about what to expect from
their students, they showed strong instructional practices. Teachers in general
showed a higher level of instructional practices but lower level of
expectations. The teachers agreed that they were effective in their
instructional practices (mean = 3.69, SD = .58), but they were not sure
(mean = 3.11, SD = .65) about what to expect from students (see Table 2). As a point of reference, the Center
for Effective Schools at the
Confidence level and external factors. Despite their low expectation level, when asked about their confidence level they indicated they were confident (mean = 4.20, SD = .70, scale, 1 ~ 5) they can change student learning in positive ways. When asked about external factors that would affect their confidence level (survey question: “To what extent do/will factors beyond their control influence their confidence to impact student learning,?”) they indicated that in a few aspects they felt there were external factors beyond their control that affect their confidence level in increasing student learning.
Expectations and instructional practices were both significantly correlated with teachers’ confidence level (r = .438, p<.01; r =.344, p<.01, respectively) (see Table 3). Teachers with higher levels of confidence also showed higher levels of expectations and instructional practices. On the other hand teachers who lacked confidence showed lower expectations for their students’ academic achievement.
Their sense of control over external factors that might influence their confidence level was not correlated significantly with Expectations, Instructional practices or Efficacy measure.
Religiosity
was negatively correlated with efficacy level (-.327*) which indicated that
efficacy level was higher for less religious teachers (indicated by less number
of times they attended religious services).
Religiosity was not significantly correlated with any other efficacy
measures. Although the number of years of teaching experience was significantly
related to the age of the teachers, as expected, teaching experience was not
significantly correlated with any efficacy measures.
Table 4 compares mean efficacy measure by grade level of the teachers. There were no significant mean differences in the expectations and instructional practices among the three low achieving schools or by grade level: primary, intermediate and upper grade. However, upper grade teachers tended to say factors beyond their control influenced their confidence to impact student learning more than the primary and intermediate levels, but not significantly.
Primary grade teachers tended to show the highest level of efficacy level. However, intermediate grade teachers showed the highest level of instructional practices and the lowest level of expectations from their students. Upper grade teachers showed the opposite; compared to other grade levels the upper grade teachers showed the highest level of expectations with the lowest level of instructional practices (see Table 4).
Efficacy level by teachers’ background. The results of mean comparisons of efficacy measures by teachers’ background measures are shown in Table 5. There were no differences in the efficacy level by gender. However, female teachers tended to show a higher efficacy level on average than male teachers. They also showed higher confidence level, a sense of control, and expectations than male teachers, while male teachers showed a higher mean on instructional practices. Despite these tendencies, there was no statistically significant difference between male and female teachers.
In terms of teachers’ educational level, those who earned a masters degree tended to show higher means in all efficacy measures than those with a bachelors degree. There were no mean differences in efficacy measures between teachers who indicated they have a religion or not, or between small class size (20 or less number of students) and large class size (21-35 students in the class).
Latino teachers showed a significantly higher level of expectations towards the predominantly Latino student populations than White teachers (p < .035). Latino teachers also showed higher means, though not significantly, in all other efficacy measures. Married teachers showed a significantly higher sense of control than single teachers (p < .063). Married teachers also tended to show a higher mean in their instructional practices, though not significantly higher than their counterpart.
Peer expectations. The teachers were asked about their expectations of the students as compared to their peers. Teachers’ own expectations for students’ potential for learning were compared to what teachers’ beliefs were of other teachers’ expectations. The results showed that while 74.6% of the teachers agreed or strongly agreed that students are capable of mastering grade level academic objectives, only 65.6% agreed or strongly agreed that teachers at the respective schools in general believe in students’ potential.
The teachers’ expectations for their own students’ academic performance were compared to their expectations for students in general. While 48.4% of the teachers agreed or strongly agreed that nearly all of their students will be at or above the national average, only 27.5% agreed or strongly agreed that students in general will perform above the national average [X2 (16, N =85) = 46.74, p<.001). The teachers had significantly higher academic expectations of their own students than those of others.
Qualitative
analysis
Confidence level. Figures 1 and 2 display the themes that were prominent in the teachers’ answers to qualitative questions that help to explain why they rated their levels of confidence as they did, and what external factors influenced their confidence to impact student learning. Boxes were drawn to show the themes most commonly mentioned and the amount of overlaps among the different themes. Seventy percent of the teachers who responded to the question indicated the main reasons that affected their high level of confidence was “teachers’ experience” and “influence on students.” Teachers believed that regardless of students’ test scores or outcomes, if they believed that students were learning, it was a positive experience, that in turn gave them confidence. This main source of confidence was followed by other reasons, clustered into the order of indication following: “teachers’ self-assessment, reflection and opportunities to role-model, and teachers’ high expectations of students.
The main factor that affected their low confidence level was teachers’ stress, due to external factors such as lack of administrative support and lack of time for individual attention to students, implementation of different “useful” programs, and perceived limits to affect achievement.
Some of the comments given by teachers as the source of their confidence were:
“I
pay attention to students’ learning, to how and what they learn and I pay
attention to the models of instruction and the learners’ experiences I provide
for children. In these ways I observe
what works and what doesn’t, then I adjust in order to make a positive effect”.
“Because I feel that I can reach the
students in a positive way, I can relate to them and sometimes they can relate
to me. Communication is a big part. Also I make my lessons humorous. Humor is a big part in keeping children’s
attention”.
“Because I am a good role model for my
students, they respect me, and most importantly they know that I care.”
“When I see that my students are learning
I’m very encouraged to push them more because I know they are ready for it and
they are building confidence, but I haven’t seen that too much this year so I’m
trying to stay confident in my ability, but I don’t really know how effective
I’ll be”.
“I am not completely confident because I
feel I am a work in progress. I have a lot to learn but I feel that the
teaching practices that I have learned have had a positive impact on my
students’ learning”.
“I’m enthusiastic about seeing my students
get involved with all learning activities. I try to prepare and present work that
is fun and applicable to my students’ lives and development. Not every concept
is grasped by every student every time but I feel my students this year enjoy
coming to school and learning in my class.”
“There are many students that I can really
reach even though there are also many I can’t. The amount of time I can spend
with any individual student is somewhat limited and therefore limits how much
impact I can have.”
“Too stressed with all the demands. Feel alone without peer help—am older—most
teachers are younger- w/o connection”
“As a teacher, I try to continually inform myself of new strategies and techniques to teach the students”.
External factors that lower confidence level. When teachers were asked what factors beyond their control influence their confidence to impact student learning, the most common response was “lack of support and involvement by parents.” Other common themes indicated were: the effects of low socio-economic-status of the parents and poverty on students, lack of support from school administrators and demands that negatively impact time spent on teaching, perceived students’ lack of motivation, focus, and effort, and language: the effects of Spanish as the predominant language in the community.
Limitations
of the Study
There are a variety of instruments and models for measuring efficacy although the majority of studies investigating teacher efficacy have used Gibson and Dembo’s (1984) scale. The scale items contain some awkwardness in wording. This can contribute to measurement error if teachers have difficulty interpreting certain statements. A formidable obstacle exists in asking teachers’ views of culturally different learners. The question leads to socially acceptable responses. Teachers do not like to think of themselves as people whose behavior or attitudes towards others might be influenced by language, appearance or culture. We do hope to better understand relationships that may exist. The results of this study will be useful for similar schools that serve low socioeconomic areas.
Conclusion/Educational Importance
As teachers’ efficacy level were explored in the low performing schools in an urban district, the results reveal that the overall levels of efficacy, expectations of students and confidence in their teaching were low.
While not tested in this study, the study’s findings are daunting in the context of previous research that established a relationship between teachers’ high level of efficacy and an increase in students’ achievement test scores (Ashton, Webb & Doda, 1982-1983). Ashton & Webb’s later study (1986) and others showed that teachers with high levels of teaching efficacy are more likely to expect that all students can learn, and to feel responsible for that learning, than are teachers with low efficacy levels.
Although these low efficacy levels were not different by the grade levels they taught, Latino teachers had significantly higher expectations than White teachers for students, the majority of whom were Latino. Their higher level of expectation may be due to most of the students being Latino themselves. This finding is consistent with Oh, Ankers, Tomyoy & Llamas’ (2004) study that showed that teachers’ ethnicity affects their view of students of their same ethnicity. The results also showed that teachers’ expectations differed significantly by the teachers’ ethnicity.
According to the qualitative survey results, teachers’ own expectations and their views of their peer’s expectations for student learning or academic achievement were different. Teachers viewed other teachers as having lower expectations than their own expectations. They indicated their own students would exhibit higher academic achievement, and students other than their own would exhibit lower academic achievement. This finding that teachers believe their students had the capacity to learn and the promise to achieve, while they were less sure of other students and of their peers, points to low collegiality and collaboration among teachers in the inner-city schools.
Many obstacles threaten teachers’ and students’ quest for educational equality. The under-achievement of minority students has prompted many studies designed to identify factors that may contribute to academic disparities. Our educational practices are clearly unsuccessful for a vast number of these students. As ethnic minorities become the majorities in public schools, we must confront the issue of how we perceive and affect the children whose hopeful faces gaze back at us, lest we and they come to expect, persist, persevere, and hope, less.
Table 1. The Two Themes from the 23-item Teacher Efficacy Indicators and Their Reliability Test Results.
|
Theme
Constructs |
Question Items |
Cronbach’s alpha |
|
Instructional Practices (Question items: 3 7 8
9 12) |
.65 |
|
|
§
When I really
try, I can get through to most students. §
If one of my
students could not do a class assignment, I would not be able to accurately
assess whether the assignment was at the correct level of difficulty §
If a student in
my class becomes disruptive and noisy, I feel assured that I know the
technique to redirect him quickly §
When the grades
of my students improve, it is usually because I found more effective teaching
approaches. §
If students have
little discipline at home, they are unlikely to accept any discipline. |
|
|
|
Expectations
(Question items: 18-23) |
.72 |
|
|
§
Most of the
students in my school will be at about the national average in academic
achievement. §
Most students in
my school are capable of mastering grade level academic objectives §
Teachers in my
school generally believe most students are able to master the basic
reading/math skills §
I expect that
most students in my school will perform at about the national average in
academic achievement §
Nearly all my
students will be at or above grade level by the end of this year §
I expect most
students in my school will perform below the national average in academic
achievement |
|
|
Note. Instructional Practices is defined
as teachers’ sense of efficacy in their instructional practices. Expectations
is defined as teachers’ expectation of students’ potential for learning.
Table 2. A Descriptive Table for Efficacy Measures (N=87, Scale: 1-5).
|
Measures |
Mean (interpretation) |
SD |
|
|
Expectations |
3.11 |
(“Not sure”) |
.65 |
|
Instructional practice |
3.69 |
(“Agree”) |
.58 |
|
Efficacy |
3.42 |
(“Agree”) |
.33 |
|
Confidence level |
4.20 |
(“Confident”) |
.70 |
|
External Factors |
3.12 |
(“In a few aspects”) |
.75 |
Note. Expectations and Instructional practices are constructs from 23 item questions on Efficacy. Efficacy is total efficacy measure of all 23 item questions averaged. Confidence level: “How confident are you that you can change learning in positive ways?” External factor: “To what extent do/will factors beyond your control influence your confidence to impact student learning?”
Table 3. Intercorrelations of Pertinent Variables
|
|
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
|
1. Confidence |
-- |
|
|
|
. |
|
|
|
|
2. External factor |
-.029 |
-- |
|
|
|
|
|
|
|
3. Expectation |
.438** |
.133 |
-- |
|
|
|
|
|
|
4. Instructional
practices |
.344** |
-.006 |
-.102 |
-- |
|
|
|
|
|
5. Efficacy total |
.514** |
.066 |
.621** |
.483** |
-- |
|
|
|
|
6. Religiosity |
.002 |
.152 |
-.108 |
-.128 |
-.327* |
-- |
|
|
|
7. Years of teaching |
-.048 |
-.081 |
-.052 |
.049 |
-.093 |
.082 |
-- |
|
|
8. Age |
-.056 |
-.165 |
-.090 |
.035 |
-.249 |
.276* |
.826* |
-- |
Note.
Religiosity measure was based on a survey question: how often do you attend
religious services.
**p<.10, *p<.05.
Table 4. Means Comparison of Efficacy Measures by Grade Level of Teachers in All Three Schools.
|
Efficacy Measures |
|
Primary Grades (n = 23) |
|
Intermediate
Grades (n = 34) |
|
Upper Grades (n = 30) |
||||
|
|
|
Mean |
SD |
|
Mean |
SD |
|
Mean |
SD |
|
|
Confidence Level |
|
4.12 |
.711 |
|
4.29 |
.611 |
|
4.33 |
.778 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
External factor |
|
3.08 |
.702 |
|
3.08 |
.669 |
|
3.17 |
.835 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Expectation |
|
3.07 |
.410 |
|
2.82 |
.423 |
|
3.17 |
.947 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Instructional Practices |
|
3.62 |
.586 |
|
3.81 |
.563 |
|
3.58 |
.802 |
|
|
|
||||||||||