smith j. ritson and richard j. robert - human · pdf filegiven training which included...

Interaction analysis

f the 1 in-

Student and Teacher Interaction Analysis: A Comparison of Activities, Age Groups and Sex of the Students in

Physical Education

oring dents

Robert J. Ritson and Richard J. Smith fsical University of Oregon Ohio and rsity

n the isser- Ann

Hughie I. Twa

Sparwood High School Sparwood, British Columbia

icro-

In the context of the classroom, the com- that through process instrumentation or plexity of interactions between and among systematic observation, scientific theories students, teachers, and curriculum has become a well discussed, debated and researched topic. Analysis techniques range in construction from result, or product testing, to a moment-by-moment account- ing for the process interpretation of the ob-

able behavior episodes and patterns registered in the classroom proceedings. Clearly, both types of studies are valuable and necessary if a more lucid understand- ing is to be established in the direction of a more effective learning environment.

According to Locke (19771, physical lcators have treated the gymnasium as a

are box" where students, teachers and

an la have been placed'in the gym for a of time. In short, effective teaching is

IS in died asmed by student accomplishment. li.e.,

end justifies the means). This black box nomenon makes no attempt to describe process in learning. Locke concluded the field of physical education must

and scrutinize the contents. st to more clearly understand ducation class, Cheffers (1977)

that the process must be under- ical education is "to.be dragged

re of myth and egocentric " (p. 26). Cheffers indicated

will eventuallv evolve which will lead to valid assumptions about teaching and learning.

Withal (1949) and Flanders (1960) produced the most widely recognized and frequently used systems for describing and analyzing the teacher/learner interaction pattern. Amidon and Simon (19651, Dougherty (1970, 19711 then Amidon et al. (19761, added categories of expected episodes of meaningful nonverbal behavior. This nonverbal category of interaction was a major breakthrough for the physical educator because activities in the gym, like developmental exercises, motor skill practice, and even creative movements, could be quantified as productive on-task behaviors. Dougherty, on behalf of the physical education research community, questioned Flander's assumption that the verbal behavior of an individual is an adequate sample of the individual's total behavior.

Physical educators and physical education students demonstrate interesting andmique patterns of interactions, as noted by Anderson (19751. I t was observed that the interchange seemed to allow students lit- tle time to give the teacher verbal feedback. "Teacher-talk, student-listen, student-

Ritson, Smith and Twa

move, and teacher-watch" appeared to be the general sequence, or pattern of interaction recognized in the gym (p. 34). The regular classroom setting seemed to follow a pattern of "teacher-talk, student-listen, followed by student-talk, and teacher- listen" (p. 34). Anderson concluded that interaction behaviors were different in physical education than regular classroom activity.

Previous research has shown that a clear difference exists between the regular classroom and the gymnasium interaction patterns of students and their teachers. The purpose of the present study was to in- vestigate interaction of teachers and learners in physical education at an elementary school age level. The study examined the variables of age level of the student groups, sex of the student groups and the activity.

Cheffers expanded Flander's instrument to 20 interaction categories so it could be appropriate for use in the physical education setting (1972). I t was later published by Cheffers, Amidon and Rodgers (1974) entitled Cheffers' Adap- tation of Flander's Interaction Analysis System (CAFIAS). The instrument contained 14 categories of teacher interaction and six categories of student interaction. Research by Cheffers, Mancini and Mar- tinek (1980) concluded that few differences existed between elementary and secondary level classes or sex of the teacher when usage of the categories, interaction parameters, and patterns of interactions were observed. Teachers seemed to use lecture and direction as predominant style of contacting the students. Specific categories that yielded significant results were: student initiated nonverbal responses--very few noted; punishment/correction of students--low frequency; teacher's acceptance of student ideas--low frequency; and teacher's praise-- low frequency.

Rankin (1975) modified CAFIAS for use in classroom situations and developed a system for observing the interaction patterns of elementary school physical education classes that included ten categories of in-

teraction behavior (five verbal and five non verbal behaviors). The instrument wa validated by collapsing the CAFIA! categories which related directly to th Rankin Interaction Analysis Systen (RIAS). Rankin's criterion referenc validity coefficients of r l = .72, r2 = .6! and r3 = .62 were accepted as recommend ed by Dayton (1970). These coefficient. represent inter-rater agreement using the in terval-by-interval(1-I) system.

Rankin concluded after 42 observation that male and female instructors differed i~ their use of one nonverbal behavior namely, teacher gestures. Between primay level children and intermediate level subjec observations, more students displayec frowning and displeasure for the physica education sessions at the intermediate leve (grades 4-01 than at the primary level (K-3) Rankin alluded to but made no attempt tc study relationships among the curriculal areas of physical education. Rankir suggested that the type of activity in tht lesson may be important to consider in tht interpretation of the interaction data. In 2

follow-up study by Twa (19791, activity war again mentioned but no attempt was madc to control this variable; however, it wa! recommended that further study with con, trol of activity and sameness of lesson ir warranted.

In summary, more diverse projects witk better attention to confounding variable! (e.g., experience of teacher, length of class. activity, skill level, environment. and man) others) should be instituted to investigatt verbal and nonverbal interaction.

Instrumentation

The Rankin Interaction Analysis System (RIAS) was selected as a suitable tool for this study. The RIAS (1975. 1978) is an adaptation of the systems proposed by Flan- ders (19001, which dealt with teacher and student interaction in the elementary school classroom, and Cheffers (19721 who developed a system of analysis to include

*


ations *red in avior, .imary

nonverbal interaction between teacher and students in a physical education setting. Rankin's adaptation was a simplification of Cheffer's 20-category system, while adopt- ing the Flander's method of tallying the data in coupled observations which are tabulated on a 10 x 10 matrix to show linkages and sequences of the observed interaction behaviors (1975).

In the present study, it was necessary to clarify Rankin's definitions of the 10 interaction categories. They were established as follows:

1) Teacher Talk Talk by the teacher in giving directions commands, or orders, lecturing, ex- pressing ideas, or answering questions;

2) Teacher Rejection The use of statements by the teacher intended to change or control negative student behavior;

Talking with classmates;

Talk when asking questions, or answering questions, or comments directed by

Positive verbal acceptance or reinforce-

, with iables class, many tigate

ment of student behavior;

reacting positively to com- instructions from the teacher

s smiling, laughing, ggling or facial approval;

y student moving that is a result teacher talk which includes lining

des frowning and

9) Teacher Gestures Major pedagogically-related gestures that support or aid teacher talk, teacher rejection, or teaching praise. For example: pointing, holding up a certain number of fingers, smiling, any hands-on communication (pats, hugs, pushes), nodding or shaking the head, spotting, demonstrating, blowing whistle, etc.;

10) Non-Response o r Confusion Confusion, chaos, disorder, noise. The students don't know what to do and are awaiting direction.

Procedures

The setting of the study was the Children's Summer Sports Program at the University of Oregon in Eugene, Oregon. The program serves children ages 8-13 with physical activities and sport skills during the summer vacation. The program runs for seven weeks, five days per week with three hours of activity each day. Children are grouped in 10 sections by age and sex. Par- ticipants are further divided into subsets according to experience in the program and special consideration (e.g., sibling, size, handicap, etc.) to establish relative homogeniety. Most children reside in the greater Eugene-Springfield, Oregon area but a small percentage come from other locations around the world whose parents are affiliated with the University's summer session. The program is longitudinal in nature, therefore, it is not unusual for a youngster to participate for up to six years. The philosophy of the program is geared toward individual skill teaching and refinement with a healthful balance of cooperative and competitive experiences for the purpose of learning specific game skills and strategies.

All subjects participated in gymnastics and swimming sessions which were 40

Riston, Smith and Twa

minutes each and met daily with 1:14 in- structor-student ratio. The majority of the instructors were master's degree and doctoral candidates with several years teaching and coaching experience. Qualified persons were selected for their demonstrated exper- tise in a specific content area and for the ability to teach the particular age groups included in the program.

The subject setting for the observation and assessment of student and teacher interaction was a gymnastics apparatus room, or a swimming pool. Teachers were inform- ed that a person would make an observation of the fist 20 to 25 minutes of class at some time in the next few weeks. Students in the classes were told nothing. Obser- vations of this nature were not uncommon and program participants rarely noticed the observers. There were no refusals among the instructors that were selected to be observed. Instructors were observed teaching in each of four graduated age brackets. A total of 12 instructors were observed (six teaching gymnastics, six in swimming). A total of 48 observations were scheduled. All observations were completed in an 11-day period during the fifth, sixth, and seventh weeks of the program.

One doctoral, one master's, and one senior undergraduate student, a11 majoring in physical education, agreed to be the data collectors for the project. Observers were given training which included reading, discussing and practicing in coding the categories of the RIAS.

The objectivity of the instrument was calculated and produced Spearman-Brown rank correlation coefficients which ranged from -749 to .935 for the six pairs of data. The coefficients were not accepted and were analyzed by the investigators.

A further meeting was held to clarify only the categories of interaction behaviors which were inconsistently interpreted by the four members of the observation team. Observers were instructed to practice and do a second objectivity observation. The second observation was made and the reliability coefficients were -978 (AB), .979 (AC), .994

(AD), .939 (BC), .961 (BD), and .979 (CD). The letter "A" above represents the trainet and the letters "B,C,DW represent the trainees. I t was concluded by the investigators that the coefficients were relatively consistent and that the project should continue to into the data gathering phase.

, Finally, the observers were given a schedule of observations to complete in the following two weeks. They were left to decide which class would be observed and when. Each observer was scheduled with an equal number of observations. The observers completed their portion of the project by meeting with the primary investigator for an interview and discussion following the data collection.

Results and Discussion

Of the 48 scheduled observations, M were completed. Four observations were canceled due to unnsual circumstances. In all, 22 observations were completed in each subject area, gymnastics and swimming.

For the purposes of reporting the data and findings, a three-pronged approach was used. First, descriptive data was provided of all observations (Table 1) with a summary of the categories of interaction and the amount observed. Secondly, by employing the chi square statistical technique category by category and isolating the independent variable (sex of student, age group, and activity), differences among the categories were tested for significance. Finally, the categories of interaction which were recognized to have significant difference5 were analyzed by the use of the visual inspection technique of averages.

During the 44 completed observation: there 16,842 interactions tallied betweer the teacher and the student groups which produced a mean number of interac. tions per observation of 386.5 in a 20-minutt period. The observers were trained to recorc interactions at an approximate three-seconc interval as recommended by Flander! (19601.

Table 1 Descriptive Data from All Observations by Interaction Categories

Interaction Gymnastics Swimming Male Female Age Group Total % of Category Frequency Frequency Frequency Frequency I I1 I11 N Frequency Total

(N.22) (N-22) (N.23) (N=21) (N.12) (N=l l ) (N=ll) (N.10) (N.44) Frequency

1. Teacher Talk 2. Teacher Rejection 3. Student Talk 4. Student Feedback 5. Teacher Praise 6. Students Smiling 7. Students Moving 8. Students Frowning 9. Teacher Gestures

10. Non-Response or Confusion

An observation ref- to class session observed. A frequency refers to a noted occurrence within an observation.


Teacher talking, students moving and teacher gesturing were clearly the most numerous interactions noted by the observers (see Table 1). Student feedback to the teacher was also observable with a relatively consistent frequency from observation to observation. Teacher rejection, students displaying displeasure (frowning) and confusion were almost never observed.

Since the four age groups, sex of the participants and activities were independent and the data were in terms of frequencies in discrete categories, the chi square test was selected as the appropriate statistical procedure for analyzing the data. Because differences did exist, it was necessary to do a complete break-down of all of the statistical analyses. These can be seen in Tables 2, 3, and 4.

Table 2 contains the results of the chi square analysis between the categories of interaction and the activities of swimming and

gymnastics. The results indicated that there were significant differences in the area of teacher talk. The data analysis indicated that the gymnastics classes contained a higher frequency of teacher's verbal instruction. The other nine categories tested indicated no significant differences.

The four age groups, when sex and activity were not separated, were tested for differences. The results indicated that two categories demonstrated significant differences. Students smiling and students moving ( p < .05) showed a difference wor- thy of further investigation.

When combining age groups and activities, thus separating out the sex of the participants, the results of the chi square analysis can be seen in Table 3. Students smiling, especially the female groups, can again be seen to show a significant result along with the category containing the frequencies of students moving (p < .05).

TABLE 2 Chi Square Test of Significance Between Interaction

Categories and Activities

Categories of Interaction X X

Gymnastics Swimming X

1. Teacher Talk 2. Teacher Rejection 3. Student Talk 4. Student Feedback 6. Teacher Praise 6. Students Smiling 7. Students Moving 8. Students Frowning 9. Teacher Gestures

10. Confusion or no interaction

there ea of cated

1 ac- 1 for . two

dif - ients wor-

ac- l the 'curre bnts

can esult

the ;I.



Categories on Sex of Students

X X egories of Interaction Male Female X

(N=23) (N=21)

1. Teacher Talk 131.13 111.80 1.54 2, Teacher Rejection 1.57 0.57 0.21 3. Student Talk 12.61 6.05 2.27 4. Student Feedback 23.78 34.38 1.95 5. Teacher Praise 9.35 7.81 0.13 6. Students Smiling 0.96 9.95 7.72** 7. Students Moving 131.10 99.14 5.74* 8. Students Frowning 0.35 0.05 0.23 9. Teacher Gestures 90.74 85.95 0.13 10. Confusion or no interaction 0.61 0.71 0.01


Categories and Age Groups a

X X X X Group I Group I1 Group I11 Group IV (N=12) (N=ll) (N=ll) (N=10) X

~eacher Talk 136.00 119.36 123.00 106.60 3.61

o** 0 0 0 0 107.92 139.54 96.54 132.10 10.31* 9 8 06 5 8 - 5, df = 3 a Children wereaclassified by age and sex (e.g., Group I was males and

in separate sections with a different teacher but all children were 8 and 9 years old).


Of the original ten categories of interad- tion behavior, there were three categories that when tested indicated significant differences: teacher talk, students smiling, and students moving. A visual inspection technique was employed to aid in the for- mulation of conclusions from the research data as well as suggest avenues for further experimentation. The statistical moving averages reported in Figures 1, 2, and 3 represent the mean data from the total of 44 group observations. Each figure contains the same data aggregated in two ways (age groups and sex of the student groups).

Interaction behavior category #1 (teacher talk) is illustrated in Figure 1. I t seems clear that the male gymnastics classes demonstrated a consistent high frequency of teacher talking from age group to age group. The female gymnastics classes, the male and female swimming classes, although somewhat dispersed, showed a regression downward in the amount that the teacher employed verbal episodes. When

the males and females data (age groups I , 11,111, IV) were aggregated, the amount of teacher talk from swimming to gymnastics differed widely (especially in the male data).

Gymnastics and swimming require a good deal of verbal instruction. Teachers of gymnastics may verbally instruct the entire time of student contact. Swimming usually limits teacher talk to the time when students' heads are above the water. I t may also be interesting to note that teacher gestures were not measured as significantly different in the different age groups, sex, or the activities.

Students smiling, category #6, provided some predictable information as well as a problem of the instrument used. Figure 2 shows a regression downward in the number of smiling behaviors as the age increased in the student groups. The female gymnastics

.observations of age group 111 seemed to defy the overall trend. The high rate of smiling female gymnasts may be the result of curricular training in feminine gymnastics where the aesthetics, pleasure of per-

Female gymnastics

Male swimming

Female swimning

Fig. 1-Comparisons of Teacher Talk When Separated by -Age Group, Activity and Sex

22/Winter, 1982

I I I I I I t 1 2 3 4 Male Fema 1 e

AGE GROUPS

oups I, ount of mastics !data).

a good )f gym- re time y limits udents' o be in- 2s were tin the ,ties. Sovided

as a

w e 2 lumber ased in nastics ned to ate of L result

gym- of per-


2 7 - 8 I I

1 I 1 I

24 - I L I I

1 I 21 - Female gymnastics 4 I 1

I I \ I I \

E l 8 - I I Gymnastics

.. \ +

I a

\ >

I I I

4 L

\ I I I

2 1 5 - \ a

I I I

\ I I 0 \ L

I I I I

E 1 2 - I

Female swimmlng I I

1 9 - I

I I

6 - I Male swimming

I I

3 - Male gymnastics 0--- - -

I I I I 1 2 3 4 Male Female

AGE GROUPS

Fig. 2--Comparison of Student Smiling When Separated by Age Group, Activity and Sex

Ritmn, Smith and Twa

formance, and other techniques were employed to gain approval for a gymnastics performance. The interaction analysis instrument employed in the present research cannot ascertain whether the data were reflective of student approval, gymnastics skill, or experimental chance.

In category #7, students moving, a significant difference resulted and is illustrated in Figure 3. The male classes seemed to move more than their female counterparts in both gymnastics and swimming, but particularly in swimming. Gym- nastics illustrates a more stable age group graph with the exception of the female group 11. In swimming, there is a disparity among the groups which is difficult to ex- plain.

The above study seems to concur with Cheffers' (1979) summary statement that there seems to be a 60-40 relationship between the teacher and student initiated interaction behaviors (i.e., 60% teacher initiation, 40% student initiated interaction). Cheffers recorded that there were minimal differences between male and female teachers in interaction patterns. This conclusion conflicts somewhat with Rankin (1975) who suggested differences existed in the areas of student frowning and in- terestingly, teacher gestures. Twa 11979) re~orted that differences were demonstrated between male and female teachers in the areas of confusion, teacher praise, and students frowning. A comparison with Twa and Rankin is reported below:

1. Twa (1979) found that teachers gesture ap- proximately 12.4% of class time whiie in the present study the teachers gestured 23.8% of the time. It may be that the philosophy of the sports program encourages a modeling type of teacher behavior while the public school physical education classes employ predominate-

ly a verbal type teacher strategy;

2. The frequencies noted in the present study in the areas of confusion, and students frowning were extremely low and nonexistent from observation to observation, which may be a result of the volun- teer subject group or teacher selection; and

3. Twa reported that students talked 10.4% of class time in the public schools while the sports program identified students talking 2.5% of class time. The sports program may encourage less student talk than the co-ed classes Twa and Rankin observed.

Student and teacher interaction analysir with its many categories, is vital to the un derstanding of the learning situation. It ma be that future studies are needed in th student moving category alone to investigat the kinds of movements which are observ able and whether certain types o movement tend to elicit specific types of in teractions. The RIAS, as an observatiol system was adequate to answer thl questions of the present study and is ap plicable in the observation of regula physical education, student teachers, peel teaching situations and may be appropriate for limited evaluative decisions.

References

Amidon, E.J., & Hough, J.B. (Eds.). In. teraction analysis: Theory, research, and application. Reading, Mass.: Ad. dison-Wesley, 1976.

Amidon, E.J., & Simon, A. Teacher-pupil interaction. Review of Educational Research, XXXV, No. 2, 1965, pp. 130-39.

derson, W.G. Videotape databank. Journal of Health, Physical Education and Recreation, 46, No. 4,1975, PP-

strument designed to expand the Flan- ders system of interaction analysis to describe nonverbal interaction, different varieties of teacher behavior and pupil responses. Unpublished doctoral dissertation, Temple University, 1972.

Cheffers, J.T.F. Observing teaching

Cheffers, John F., Amidon, Edmund, & Rodgers, Kenneth. Interaction analysis: An application to nonverbal activity. St. Paul, Minnesota: Paul S. Amidon & Assoc., Inc., 1974.

Cheffers, J.T.F., Mancini, V.H., & Mar- tinek, T. Interaction analysis: An application to nonverbal activity (2nd ed.). St. Paul, Minnesota: Paul S. Amidon & Associates, 1980.

. Mitchell. The design of educational experiments. New York: McGraw-Hill Book Co., 1970, pp. 67-

individual programs styles on teaching in the deyelopment of physical fitness and motor skills. Unpublished doctoral dissertation, Temple Univer-

N.J. A plan for the analysis of teacher-pupil interaction in physical education classes. Quest, XV, 1971,

classroom: A manual for observers. Minneapolis: University of Minnesota,

University of Oregon, 1971. ,


Rankiin, K.D. Verbal and nonverbal interaction analysis of student teachers with students in elementary physical education. Unpublished doctoral dissertation, University of Kansas, 1975.

Rankin, K.D. An objective approach to student teacher evaluation. The Physical Educator, 1978, pp. 43-46.

Twa, H.I. A comparison of male and female physical education teachers' verbal and nonverbal interaction at the elementary school level. Unpublished doctoral dissertation, University of Oregon, 1979.

Withall, J. The development of a technique for the measurement of social- emotional climate in the classroom. Journal of Experimental Education,

. XVII, 1949, pp. 347-61.

Robert Ritson is an Assistant Professor in the Department of Physical Education at the University of Oregon, Eugene, OR 974.03. His areas of specialization are teaching methodology and curriculum at the secondary level and research on teaching at the elementary and secondary levels.

Richard Smith is an Associate Projessor in the Department of Physical Education at the University of Oregon, Eugene, OR 974.03. He is the Coordinator of the undergraduate Teacher Preparation Program, and the coordinator of the graduate Administration Program Area.

Hughie Twa is a teacher and Program Administrator in Physical Education at Spanoood High School, Sparwood, British Columbia.

smith j. ritson and richard j. robert - human · pdf filegiven training which included...

Documents