Chris Murray
Tualatin High School

Dr. Ken Peterson
Portland State University

Table Of Contents:
Abstract
Introduction

• The Research Question
• Review of Related Literature

• The setting
• Subjects
• Data Gathering and Instrumentation
• The Treatment
• Administration of tests and treatment
• Confidentiality and anonymity

The Data

• SAI for Boys
• SAI For Girls
• GWI for Boys and girls
• The class discussions
• Discussion

Abstract:  Table of Contents

The discrepancy between the number of men and the number of women who choose careers in the physical sciences is both startling and a cause for national concern. In 1960, 4% of all physicists were women. (Rossi, 1965) In 1982, more than 20 years later, the figure remained at fewer than 5% (Max, 1982). As global competition in technology becomes more and more important to our economy, we become less able as a nation to write off nearly half of our potential physicists. As a teacher of young physicists, I am aware of this trend in my classroom. We have an IB program in which the choice in science is between physics and biology. Men mostly choose physics, and women biology.

It has been my experience, and the experience of nearly every other science teacher I have met, that young women are very under-represented in upper level high school science classes, particularly classes in the physical sciences, and most particularly, physics. Researchers (Ethington & Wolfle, 1988; DeBoer, 1984; Ware & Lee, 1988) have shown a strong positive relationship between choosing a science career or college major and the number and type of high school science courses a person has taken. Many researchers (Berryman, 1983; Campbell & McCabe, 1982; Thomas, 1984) have shown that women take significantly fewer high school science courses than men. Since there is a strong relationship between achievement in college science and the high school science courses a person has taken, (DeBoer, 1984) high school seems to be a crucial time in the formation of a career choice.

Specifically, Remick and Miller (1978) have identified high school physics as a necessary prerequisite to many fields in science; students not taking physics in high school significantly limit their science options in college. Unfortunately, Smith and Walker (1985) report that the ratio of males to females in high school physics remains fairly stable at 2 to 1 despite advances in the number of women employed as engineers and scientists. It seems important to me then, to investigate further the reasons why more young women do not take physics in high school.

My own personal goal is to teach as much physics to as many people as I can, and at the same time to instill the love I have of physics in my students; I want to make them more likely to choose a career in the sciences. Try as hard as I might by recruiting, I find that the ratio of women to men in my classes stays constant at somewhat more than one to two. Somewhere out there, and not taking my classes, is a significant portion of the high school population (young women) that are able but not willing to take physics. Interesting them in physics is my goal. At our high school we now have eight sections of physics. If the classes were an equal mix of men to women, we would be offering eleven or twelve sections.

The Research Question   Table of Contents

Several years ago, I took a couple of courses on using cooperative education in the high school classroom. I had barely used any of the skills I learned, and I was wondering if using cooperative education would increase the interest of high school women in taking physics. More specifically, I was interested in the affective domain. Students tend to take courses that other people have good feelings toward, so I thought it would follow that if female students had good feelings toward physics, they would tell other female students, making it more likely that they would take the course.

The purpose of my research was to find out the relationship, if any, between using cooperative learning strategies and the attitudes toward science of high school women taking physics. I had the hypothesis that there was a positive relationship because most of what I had read about cooperative learning showed that women respond positively to it.

Cooperative learning I defined as work in small groups, with the teaching of social skills, member interdependence, individual accountability, and a common goal. I learned various strategies in my course at PSU, among them study groupings, jigsawing, and laboratory roles such as "Materials manager" and "Record keeper".

Attitudes toward science I defined as the student feeling that she is well prepared to do science, that science has made good contributions to society, and that she is better off because of science.

Review of Related Literature   Table of Contents

Starting at a very early age, girls tend to have markedly different attitudes toward science and science-related careers than do boys. Career sex-stereotypes are present as early as kindergarten (Davis, Jones & Myers, 1989; Schlossberg & Goodman, 1982) and first or second grade (Siegel, 1983; Vockell and Lobonc, 1981). Girls’ interest in science, however, keeps up with boys’ until eighth or ninth grade when boys become significantly more positive toward science than girls (Hardin & Dede, 1985). Young women in high school science consistently have lower self confidence ratings than do the their male peers (Post-Kammer & Smith, 1986). This is a concern as it seems to be a cyclical process; women who perceive themselves to be a "deviant minority" in high school science classes achieve at a lower level than women in all-female classes (Vockell & Lobonc, 1981), which in turn may make them less likely (Hollinger, 1983) to take further science courses in high school, and then ultimately it becomes a career choice. Longitudinal studies (Post-Kammer & Perrone, 1983) of talented and gifted female science students show that they feel they are living up to their occupational potential much less than the boys in their program.

What, then, accounts for young women’s lack of interest in science as a whole and specifically the physical sciences? There seems to be little conclusive evidence (Linn & Hyde, 1989) that there is an innate cognitive difference that makes young men better at physics than young women. This is supported by young women scoring (Smith & Walker, 1985) as high as young men in high school physics exams across the country. Harding and Sutoris (1984) feel that boys and girls have a different self-development. Young women are more socially oriented and get more satisfaction from working with other people in groups (Lips, 1992), whereas young men are more satisfied working by themselves, and are more object-oriented. In Western culture, then, Science has a tradition of being an individualistic asocial discipline

Young women who do excel in the physical sciences tend to adapt to their environment by manifesting the more asocial masculine behaviors of autonomy and self-reliance (Baker, 1984) while at the same time rating science as a whole as more social than do their male colleagues (Lips, 1992).

Perhaps it would be better to adapt the science classroom itself to the needs of high school women than to require the opposite. Cooperative learning (Johnson & Johnson, 1987) is one way to organize a classroom in a more social and group-oriented manner. A number of studies (Tlusty, 1993; Davis et al, 1989; Conwell, 1988; Okebukola, 1986) have shown that women achieve at a higher level in, and have a better attitude toward science, when it is taught using cooperative or collaborative groupings in class and in the laboratory. Okebukola (1986) studied the attitudes of young women in a laboratory setting. He chose two schools about 80 kilometers apart with similar populations. In the experimental school the instructors use cooperative team learning and the control school used traditional individualistic instruction. He administered a pre and post attitudinal survey to both groups, and found that women have less positive attitudes than the men regarding laboratory work when it is individually done, but have a better attitude when it is done in cooperative groups.

A recent study (Tlusty, 1993), one that is quite similar to this study, looked at the attitudes and achievement of college chemistry students. The researchers had two different classes taught by the same professor. In one class the professor used cooperative learning strategies, and in the other, more traditional instructional methods. The researchers had the students in both groups complete an attitude survey at the beginning and the end of an eight-week period. The subjects also did journal entries, interviews, and reflective writings. After the eight-week period, they switched groups, and the control group received the treatment (Cooperative learning), and the experimental group received the non-treatment (Traditional teaching). They found that college women learning cooperatively not only achieved at a higher level than the control group women learning in a more traditional setting, but also had a more positive attitude toward science and science-related careers.

Cooperative learning may be a way, then, to better adapt science as a field of study to the needs of young women to be more socially oriented. Rather than have them adapt themselves to an asocial curriculum, in ways that may damage their self-confidence and self-esteem, perhaps we should try making the environment more conducive to their preferred mode of learning.

Since there was already a precedent in the literature for Cooperative Learning increasing the achievement and attitudes of young women in college chemistry classes, I thought it was likely to have the same effect in high school physics.

Methodology  Table of Contents

The Setting   Table of Contents

Tualatin high school is a mostly white suburban school of 1500. About 60 of those students are Hispanic, and about 20 are of other race or ethnicity. Tualatin is a fairly affluent and conservative community of 20,000. My students are generally pretty motivated and academically oriented. The classes that I studied were students from four already existing first year physics classes.

Subjects  Table of Contents

The experimental and control group subjects were juniors or seniors in high school enrolled in the first year of either IB or general (G) physics at Tualatin High School. The classes were typically about 20-28 students each, with a male/female ratio of two to one. They ranged in age from 16 to 19 years in age.

I divided the four classes into two groups of two classes, each comprised of a general level (G) and IB level class, such that there were a nearly equal number of subjects in each group. I also tried to balance the male to female ratio. Below is a summary of the students who participated in the study:

Due to absences, a significant number of these students did not complete all of the surveys, and so the numbers used for statistical analysis are somewhat lower.

Data Gathering and Instrumentation   Table of Contents

For my research, I measured the constructs of attitudes toward science, attitudes toward group work, and overall general reaction. For this, I chose the Science Attitude Inventory, and a combination of class discussion and an instrument of my own devising that I called the "Group Work Inventory."

The Science Attitude Inventory (SAI)

First, I divided the class into groups of three or four. To allow the students to pick their own groups and ensure heterogeneity, I took each class and sorted the students into four groups according to test scores. I gave the students in the highest group a red card, the students in the middle two groups a blue card, and the students in the lowest group a green card. I then told them to pick groups of three or four with the rule being that there could be exactly one red and one green card, and one or two blue cards in each group. I then arranged the seating chart so that the members of each group sat near each other.

I employed a variety of cooperative learning techniques in the classes that were the receiving the treatment as part of an experimental group.

Numbered heads together:

This research used a pre- mid- and post-test design during the second and third quarters of the 1998-1999 school year. At the beginning of quarter 2 I administered the SAI and the GWI (Group work inventory) to both group A and group B. I then administered cooperative learning for quarter 2 to group A, and group B did not use cooperative learning. At the end of quarter 2, I administered the SAI and GWI to both groups, and held a class discussion with the experimental group. For quarter 3 I switched the experimental and control groups, administering the treatment of cooperative learning to group B, and group A went back to non-team learning. Finally, at the end of quarter 3, I administered the SAI and GWI to both groups, and held class discussions this time with group B.

This is summarized in the table below:
 

	Group A	Group B
Quarter 2:	Administer SAI, GWI (Pre) Cooperative Learning Treatment (Experimental) Administer SAI, GWI (Mid) Class discussion	Administer SAI, GWI (Pre) No Special Treatment (Control)   Administer SAI, GWI (Mid)
Quarter 3:	No Special Treatment (Control)   Administer SAI, GWI (Post)	Cooperative Learning Treatment (Experimental) Administer SAI, GWI (Post) Class discussion

Confidentiality and Anonymity   Table of Contents

The students identified themselves on their survey forms for the purpose of pairing scores for analysis with a confidential ID number assigned to them. Individuals were not identified in any other way.
 
 

Results and Discussion   Table of Contents

The Data   Table of Contents

The SAI and the GWI I administered on paper forms and typed the responses by hand into a very large Excel Spreadsheet. I figured out a formula to score the positive and negative items in each, and then performed paired T-tests on the results. If a student did not complete all of the Pre- Mid- and Post- tests, then they were not included in the statistics. I did not use a T-test on my group work inventory, as it is not a validated instrument. I used it for triangulation only.

In the tables below are the results of my data analysis. First on the left are the average scores for the group, then on the right are the differences, Pre to Mid, and Mid to Post. The boxes are around the change scores for the experimental groups.

Table 1 - The results of the SAI for Boys  Table of Contents
 

	Experimental				Change Scores
	Pre	Mid	Post		Exp.	Control
Period 2	178.5	171.8	169.5		-6.8	-2.3
n =	4
A				T-test:	0.319
Period 7	187.5	185.2	184.8		-2.3	-0.4
n =	14
				T-test:	0.279
				Average:	-3.3	-0.8
		Experimental
	Pre	Mid	Post		Control	Exp.
Period 4	174.8	169.1	171.6		-5.7	2.4
n =	9
B					T-test:	0.241
Period 5	171.5	175.1	165.6		3.5	-9.5
n =	13
					T-test:	0.042
				Average:	-0.2	-4.6

Table 2 - SAI Results for Girls   Table of Contents
 

	Experimental				Change Scores
	Pre	Mid	Post		Exp.	Control
Period 2	173.6	169.9	171.6		-3.7	1.7
n =	7
A				T-test:	0.171
Period 7	179.3	176.9	179.7		-2.4	2.9
n =	7
				T-test:	0.491
				Average:	-3.1	2.3
		Experimental
	Pre	Mid	Post		Control	Exp.
Period 4	175.3	173.3	176.6		-2.0	3.2
n =	9
B					T-test:	0.234
Period 5	177.3	178.4	178.7		1.1	0.3
n =	10
					T-test:	0.857
				Average:	-0.4	1.7

 

Table 3 - GWI Results for Boys and Girls.   Table of Contents
 

Boys
	Experimental				Exp.	Control
	Pre	Mid	Post		Change	Change
Period 2 - A	27.3	29.5	30.0		2.3	0.5
n =	4
Period 7 - A	27.7	29.6	30.8		1.9	1.2
n =	14
				Average	1.9	1.1
					Control	Exp.
		Experimental			Change	Change
Period 4 - B	29.3	31.0	30.1		1.8	-0.9
n =	8
Period 5 - B	27.8	27.5	29.0		-0.4	1.5
n =	13
				Average	0.4	0.6
Girls
	Experimental				Exp.	Control
	Pre	Mid	Post		Change	Change
Period 2 - A	31.9	30.7	30.7		-1.1	0.0
n =	7
Period 7 - A	29.5	29.2	31.2		-0.3	2.0
n =	6
				Average	-0.8	0.9
					Control	Exp.
		Experimental			Change	Change
Period 4 - B	26.7	27.4	29.2		0.8	1.8
n =	9
Period 5 - B	28.4	28.5	31.0		0.1	2.5
n =	10
				Average	0.4	2.2

The results of the SAI were generally inconclusive and lacking in statistical significance. Boys were generally negative as seen in table 1. In fact the only change that had less than one chance in ten of being not due to random variation was the period 5 boys reacting negatively to the treatment from mid to post. The rest of the changes ranged from -3.7 to +3.2 for the girls, and from -9.5 (significant at .042 level) to +2.4. There was no pattern of reaction that I could see between the experimental and control groups.

The GWI survey that I used had similar results. The girls ranged from -1.1 to +2.5 and the boys ranged from -0.9 to +2.3 with no real pattern either.

There did not exist any pattern, significant or not, in the scores part of the data; there was no significant reaction to the main treatment of cooperative and team learning.

The Class Discussions   Table of Contents

Below is a summary of the comments made during the class discussions. The number preceding each item is the number of people who contributed that idea or statement.

Period 2

I was surprised that there was not a significant positive reaction to the treatment according to the SAI, especially among the young women. I can think of several reasons why I did not get the results that I hypothesized and that others have obtained in their studies.

Largely, I think that since this was the first time I have taught using cooperative strategies on such a grand scale, the treatment was not effective because I lacked confidence and competence in using team learning.

Possibly the students were reacting to physics in general--a challenging course of study. My class is popular, but difficult; the students might have had a negative reaction to physics that masked a potential positive reaction to Cooperative Learning. I don't think this is too likely, but I am looking for explanations.

There might have been a negative reaction to the SAI, or I might have administered it in a negative or incorrect way. Since I have used the SAI in other studies at the University of Minnesota, I don't think this is the case either.

My hypothesis might not be correct--High School women are not actually positive to cooperative learning. This would fly in the face of almost all the literature that I have read on the subject.

Finally, it might be the case that they are already very positive about science (due no doubt to my superior instruction!) and are not going to increase much due to any treatment.

Since the scores were not significantly different, I am not worried about the general negative trend, but I am curious. I may do a pre- post-test next year to see if there is a "negative thing" going on in my classroom. I certainly do not think there is, but if there is, it is a cause of concern for me.

The results of the Group Work Inventory did not seem to be related in any way to the results of the SAI. There was no trend either way linking the two.

Despite the lack of positive results from the SAI and GWI data, the class discussions were very positive. Every class had a choice for the last quarter of the year of whether to use groups or not, and every class voted overwhelmingly to use cooperative learning. Although this was possibly just because they wanted to take group quizzes, I cannot ignore the positive things that happened in class and were said in the group discussions. The prevailing positive remark was some variation on the idea that groups provided a group of people, one of whom could always answer your questions. They liked being able to confer within their groups and work together on projects - the old adage that two heads are better than one. This is not too startling, as it is the primary advantage of using team learning. Traditionally, there is only one teacher in the classroom, and with team learning, everyone is both teacher and learner.

One remark I thought was interesting was that during whiteboard time in class, they really liked having someone else to ask questions of about the whiteboard problems. This came out in both of my IB classes independently. In both cases, I pointed out that they could have asked questions without having formally assigned groups - in fact it is something that I have always encouraged. Their reply in both cases was that having the groups formally assigned gave the members a sense of responsibility for the group, and a sense of ownership that was not present in a traditional setting. There exists, then, a need for some kind of social framework or expectations to facilitate what I thought was already happening. I would never have dreamed this would be the case had I not heard it from them myself.

Other positive comments that I thought were interesting were that cooperative groups were helpful to ESL students. This is something I intend to share with our ESL staff here at the school.

The negative remarks that were made tended to be specific and constructive in nature, dealing with items or logistical details that could be improved. To be sure there were a few holdouts who were undeniably negative. These students tended to be comprised of the truly gifted and bright students, but did not represent them in their entirety. (There were some really bright students who took well to groups.)

One source for many of the negative remarks was a group in my 7th period class that did not get along in a spectacular way. They were a constant source of amusement to the other groups with their bickering. I chose not to separate them because I felt it was important for them to resolve their differences. In retrospect, I probably should have separated them or helped mediate.

The big picture is that cooperative learning is here to stay in Tualatin High School physics. Dusting off my old notes on cooperative and team learning and actually applying them to my teaching was a hoot. I have really enjoyed using groups in my classes, and I will incorporate and improve upon my strategies. This research project has forced me to take enough data to ensure that I can improve my teaching.

Next year I am going to use groups in my first year classes, and give more group quizzes. Shorter ones, and more, as a sort of every other day check-in sort of thing. I think that will help the students make sure they are keeping up with things.

I will also continue to use the "numbered heads together" and "group share" strategies as I received a great number of positive comments for those as well.

I need to figure out a way to facilitate laboratory jigsawing. Perhaps I will make a place in the room where they can store their lab notebooks and data so that if one person is absent, the others can find the data or part of the lab that that person was working on. Jigsawing a major lab assignment is the ultimate sink-or-swim situation.

My students and I were overwhelmingly positive about team learning, and I really think it helped them learn physics. That was how I survived being a physics major; I never could solve all of the problems or understand all the concepts alone, but I could with a little help from my friends.

Acknowledgements  Table of Contents

I would like to thank Frances Lawrenz of the University of Minnesota for teaching me all about the SAI and how to do statistical analysis, Dr. Anne Wax and Dr. Stevens for showing me how to do a literature review, Dr. Dave Cox for the inspiration, and Dr. Ken Peterson for agreeing to be my action research and masters advisor.

My student assistants Pat Carlson, David Bunce, Mark Shafer, and Micah Dickerson did massive amounts of data entry.

I followed in the footsteps of Tom Duggan, who guided me through the PSU experience with sage advice.

My father, the English Professor, read through this paper and nearly ran out of red ink.

Finally I would like to thank my wife Shannon, and my children Keenan and Caelan for tolerating my night classes. What a long strange trip it's been.