Wednesday, March 21, 2012

11COURSE PARTICIPANTS’ VIEWS ABOUT TTRIBUTES RELATED TO MALE AND FEMALE TEACHERS

BY: BASHIR AHMED KHAN (PDT)

INTRODUCTION

Gender is a term used to describe socially driven aspects of our lives such as the role we play, the responsibilities we take on, the views we perceive and the behavior and activities we engage in. Among many perceptions on gender, I have observed and experienced as a student as well as a teacher that male teachers are often associated with boys and female teachers among girls in Gilgit-Baltistan, Pakistan. Various studies lend support to this view that matching with teacher and pupil by gender leads to high achievements in education. Martin and Marsh (2005) supported my view that boys fare better in classes taught by males and girls fare better in classes taught by female in subsequent Australian study. Similarly, matching teachers with students by gender has a positive impact on educational performance (Carrington, Tymms and Merrell, 2008). However, Tymms (1999) states that boys and girls are very similar in their attainment levels throughout the primary education.

Based on my observations, experiences and above facts, this study explores M.Ed[1]course participants’ (CPs) views about male teachers produce more positive attitude among boys and female teachers among girls at a private university in Karachi, Pakistan.

This paper concentrates on purpose of the study, research hypothesis, research design, sample and sampling procedure. It also discusses about measurement tools, data collection procedure, ethical consideration and limitations of the study. Data analysis, findings/results discussion and conclusion are also part of the study.

Purpose of the study


The purpose of this study is to explore the M.Ed. CPs’ views at a private university about male teachers produce more positive attitude among boys and female teachers among girls.

This study would be useful to know whether matching teachers and students by gender has any differential effects on children performance in education. It is because many studies claim that boys are underachieved due to feminization of teaching (Jha and Kelleher, 2006; Carrington and Mcphee, 2008).

Null-hypothesis of the study


There is no difference in M.Ed. CPs’ views at a private university that male teachers produce more positive attitude among boys while female teachers among girls in Karachi, Pakistan.

METHODOLOGY

Research design

In order to investigate the CPs’ views about students’ performances with matching teachers by gender, I used survey (Cross Sectional) as research design which helped me to collect information (views) directly from the selected sample once at a time. I wanted to compare male and female CPs’ views about attributes related to male and female teachers which helped me to explore students’ performance with respect to teachers by gender. I found survey is an appropriate method to use for comparison. Cohen, Monison, & Marrison, (2000) also favor it by saying that “survey method is the best method used to compare two groups to see how much they differ from each other” (p10).

Sample and sampling procedure


The sample on which this study was conducted comprised 49 CPs (24 female and 25 male). The sample belonged to different sex based sections of class M.Ed. 2011 and M.Ed. 2012 in a private university in Karachi, Pakistan. The type of sampling used for this study was non-probability sampling, more specially convenience sampling. This sample was simply available to me by its accessibility (Bryman, 2004).

Instruments/measurement tools

A questionnaire (see appendix B) was used as a major tool for data collection. The questionnaire was comprises 14 items of CPs’ views about attributes related to male and female teachers. Each item was scored on a five-point scale of (5) extremely well, (4) very well, (3) somewhat, (2) poorly and (1) not at all.

Procedure


Direct mode of administration was used as method to deliver of the instrument. I distributed the questioner among CPs and requested to fill the questionnaire. The invitation letter (see appendix A) is attached with the questionnaire. I asked the CPs to give a number which describes their level of agreement about the attributes related to male and female typical teachers. I collected the questionnaire next day when CPs filled out the tool.

Ethical consideration


To deal with all ethical considerations, I got oral consent from my participants. Similarly, I attached consent form (see appendix A) with the questionnaire and requested to sign if they are willing to participate in the study. Furthermore, I made sure the CPs that the data will not be shared with anyone else and it will only be used to an academic paper for gender in education course.


Limitations of the study

The sampling procedure was convenient sampling. Thus, the results cannot be generalized to all CPs because; this sample cannot be representative of the whole population (Bryman, 2004). As it is obvious that in quantitative study it is important to have a large sample in size but my sample size was small which is not enough to generalize the findings in any other similar context.

Reliability of the instrument

In order to find out the reliability of the instruments, I applied Cronbach Alpha test reliability for CPs’ views about attributes related to male and female teachers in table 1.1. The value of alpha α =.694 shows that the instruments were reliable. I have piloted the instruments before it administer. I found some instructions are needed to modify. After appropriate changes, I administered the tool.

Table: 1.1

Reliability of items for CPs’ views about attributes related to male and female teachers

Reliability Statistics
Cronbach's Alpha
N of Items
.694
14

DATA ANALYSIS PROCEDURE


I entered the collected data in computer and coded (e.g.1 for male, 2 for female) using SPSS[2]. It is the process of transforming a mass of raw date into tables and charts. It is also a process of artfully moulding, extracting and refining the raw data, so that the meaning and significance can be grasped. Further, in descriptive analysis, I computed the mean score of CPs’ views as independent variable and gender of the teachers as the dependent variable. Furthermore, test of normal distribution of the variables of CPs views were first tested through histogram. The histogram in figure 1 reveals that the scores about male teachers have been symmetrical distributed around the mid-point. It is perfect bell-shaped curve and shows that scores are around than the mean point.



Figure 1


Similarly, I used the thumb rule to conform the normal distribution of the data. The statistics presented in table 1.2 shows that the value of Skewness (.631) was not more than two times the value of Std. Error of Skewness (.340). This means that the data were normally distributed. Furthermore, the data was qualified for t-test for CPs views about attributes related to male teachers.

Table 1.2


Descriptives



Statistic
Std. Error
Mean_Male_Attributes
Mean
3.1035
.06511
Median
3.0000

Variance
.208

Std. Deviation
.45575

Minimum
2.21

Maximum
4.57

Range
2.36

Skewness
.631
.340
Kurtosis
.989
.668




Moreover, test of normal distribution of the variable of CPs’ views about attributes related to female teachers were first tested through histogram. The histogram in figure 2 reveals that the scores of CPs’ views about female teachers have been normally distributed around the mid-point. It is a bell-shaped curve and shows that scores are around the mean score. This shows that data is symmetrical distributed.

Figure 2


Similarly, I used the thumb rule to conform the normal distribution of the data. The statistics presented in table 1.3 shows that the value of skewness (.408) was not more than two times the value of Std. Error of Skewness (.340). This meant that the data were normally distributed. Furthermore, the data was also qualified for t-test for CPs views about attributes related to female teachers.

Table 1.3


Descriptives



Statistic
Std. Error
Mean_Female_Attributes
Mean
3.3426
.06119
Median
3.3571

Variance
.183

Std. Deviation
.42832

Minimum
2.71

Maximum
4.57

Range
1.86

Skewness
.408
.340
Kurtosis
-.066
.668



FINDINGS AND RESULTS


CPs’ views about attributes related to male teachers


The table 1.4 shows that one of the key finding from this study is that on average, the CPs views about attributes related to male teachers is average. When compared the mean scores, it is found that the male CPs scored (M=3.20; SD=.52) is slightly higher views about attributes related to male teachers than their counterpart female CPs scored (M=3.00; SD=.35). Male CPs have more dispersion from the mean. It shows that the difference within the group members of male CPs is higher than the female CPs. It means that there are male CPs who have very high and less views about attributes related to male teachers.











Table 1.4

Report
Mean_Male_Attributes


Male and female
Mean
N
Std. Deviation
Std. Error of Mean
Male
3.2029
25
.52436
.10487
Female
3.0000
24
.35308
.07207
Total
3.1035
49
.45575
.06511



Moreover, the items of the study were tested and compared individually to look for any significant difference in the scores of male and female CPs. The t-test was used to check the difference. The test statistics shown in table 1.5 that Sig.(2-tailed) values for the fourteen items are [t (47) =.118, P >.05] not significant. Thus, there is not a significant difference in CPs’ views about attributes related to male teacher.



Table 1.5

Independent Samples Test



Levene's Test for Equality of Variances
t-test for equality of means


F
Sig.
t
df
Sig. (2-tailed)


Mean_Male_Attributes
Equal variances assumed
4.338
.043
1.582
47
.120
Equal variances not assumed


1.594
42.201
.118


CPs’ views about attributes related to female teachers


Similarly, the table 1.6 depicts that another key finding from this study is that on average, the CPs’ views about attributes related to female teachers is also average. When compared the mean scores, it is found that the male CPs scored (M=3.35; SD=.43) is almost equal to their counterpart female CPs scored (M=3.33; SD=.42). Similarly, the dispersion from mean is also same in both male and female CPs. It shows that the difference within the group members of male CPs and female CPs is almost same.








Table 1.6

Report
Mean_Female_Attributes


Male and female
Mean
N
Std. Deviation
Std. Error of Mean
Male
3.3543
25
.43910
.08782
Female
3.3304
24
.42588
.08693
Total
3.3426
49
.42832
.06119



Moreover, the items of the study were tested and compared individually to look for any significant difference in the scores of CPs. The t-test was used to check the difference. The test statistics shown in table 1.7 that Sig. (2-tailed) values for the fourteen items are [t (47) =.124, P>.05] not significant. Thus, there is not a significant difference in CPs’ views about attributes related to female teachers.

Table 1.7

Independent Samples Test



Levene's Test for Equality of Variances
t-test for equality of means


F
Sig.
t
df
Sig. (2-tailed)


Mean_female_Attributes
Equal variances assumed
.103
.749
.194
47
.123
Equal variances not assumed


.194
.847
.124



DISCUSSION AND CONCLUSION


The results show that there is not a significant difference with CPs views about attributes related to male and female teachers. In other worlds, there is not any indication which shows that male teacher are particularly effective with boys or female teachers with girls. Similarly, this study also shows that there is not any indication that effective results of children are associated with male and female teachers. Furthermore, both male and female teachers are perceived as supportive by CPs. Carrington and McPhee (2008) argue that same gender matching does not have a significant influence on children achievement or attitude in education. As a result, I accepted the null-hypothesis stated early that there is no difference in views of M.Ed. CPs at a private university that male teachers produce more positive attitude among boys while female teachers among girls. However, Carrington, Tymms and Merrell (2008) claim that there is a tendency for male teachers to enhance the educational performance of boys and, conversely, for female teachers to enhance the educational performance of girls.

One reason may be that the participants of the study were adults, therefore they might have a good understanding about gender and gender related issues. They would be sensitized about gender; therefore their responses did not show any difference in attributes between male and female teachers. It may be another reason that as gender is considered a sensitive issue in our society, therefore to keep CPs themselves in a safe side they have given an average value which did not make any difference in attributes between male and female teachers. If such a study would be conducted with school children then their views might be different about teachers across gender. Thorne (1993) states that school children and adults have different perceptions about gender. Therefore, the findings of the study could be different when such a study is conducted with school children.

Second reason might be that the participants are currently studying at a private university where male and female faculties are engaged in teaching for both the gender. It may be possible that participants did not experience any difference in the performance of male and female teachers. If such a study would be conducted with single sex students with matching teacher by gender then their views would be different across gender.

REFERENCES:


Bosner, K. C. (2008). Gender Stereotype and self perction among college students. Journal of
Diversity Management. 3(3), 41-52

Bryman, A. (2004). Social research method. Oxford: Oxford University Press.

Carrington, B., & McPhee, A. (2008). Boys’ ‘underachievement’ and the feminization of

teaching. Journal of Education for Teaching. 34(2), pp. 109-120

Carrington, B., Tyumms, P., & Merrell, C. (2008). Role models, school improvement and the

‘gender gap’-do men bring out the best in boys and women the best in girls? BritishEducational Research Journal. 34(3), pp. 315-327

Cohen, L., Manison, L., &Marrison, K. (2000).Research method in education: (5thed.) New York: RuthledgeFalmer.

Jha, J., & Kelleher, F. (2006).Boys’ Underachievement in Education. Canada: Ultratech Printing

Ltd.

Martin, A. J. & Marsh, H. (2005). Motivating boys and motivating girls: does teachers’ gender

really make a difference? Australian Journal of Education. Retrieved on January 5, 2011 from http://goliath.ecnext.com/coms2/gi_0199-5013070/Motivating-boys-and-motivating-girls.html

Thorne, B. (1993). Gender play: Girls and boys in school. Buckingham: Open University Press.

Tymms, P. (1999). Baseline assessment and monitoring in primary schools: achievements,

attitudes and value-added indicators. London: David Fulton.



COURSE PARTICIPANTS’ VIEWS ON GENDER: A SURVEY

Please read the words or phrases in the first column and fill column A and B of the table below:

In column A, write a number from 1 to 5 next to each word or phrase based on how you would complete the following sentence: “This word (phrase) describes the typical male teacher…”

Extremely well =5; Very well= 4; Somewhat = 3; Poorly = 2; Not at all = 1

In column B, write a number from 1 to 5 next to each word or phrase based on how you would complete the following sentence: “This word (phrase) describes the typical female teacher…”

Extremely well =5; Very well= 4; Somewhat = 3; Poorly = 2; Not at all = 1



A
B

This word or phrase describes a typical male teacher …..
This word or phrase describes a typical female teacher ……
Helpful


Aspiring


Cooperative


Caring


Friendly


Skillful


Creative


Reserved


Assertive


Aggressive


Emotional


Authoritarian


Pessimistic


Rigid




Gender (please circle): Male Female

Class (please circle) M.Ed. 2011 M.Ed. 2012



Note: This tool is adapted from

Bosner, K. C. (2008). Gender Stereotype and self perction among college students. Journal of

Diversity Management

No comments:

Post a Comment