SOCIOMETRY IN THE CLASSROOM:

HOW TO DO IT


Lawrence W. Sherman, Ph.D.

Professor of Educational Psychology

Department of Educational Psychology

School of Education and Allied Professions

Miami University

Oxford, Ohio 45056

Phone: (513) 529-6642

FAX: (513) 529-3646

E-mail: SHERMALW@MUOHIO.EDU



INTRODUCTION

During the late 1920's and on into the 1930's two sociologists, Bogardus (1928) and Moreno (1934), because of their interests in inter-group conflict and interpersonal attraction, developed quantifiable measuring techniques which were later to be called "sociometrics." Bogardus and Moreno were primarily interested in how recent immigrants to the United States adapted to their new environment and became accepted by other non-immigrant citizens. Other researchers such as Sheriff (1936) had similar interests in the psychology of social norms. Kurt Lewin's (1931) interests in "group climates" led him to develop certain experimental methods (Lewin, Lippitt, &;White, 1930). For further insight into this interesting period of development, Peter Renshaw (1931) has provided one of the best overviews of these historical trends in his article "The roots of Peer Interaction Research: A Historical Analysis of the 1930's, p. 1-25.)

Many of these research techniques were quickly adapted to public school classroom usage and by the end of the 1950's two important books became the primary references for those interested in the classroom group process, Fritz Redl and William Wattenberg (1959) devoted an entire chapter of their classic book Mental Hygiene In Teaching, to "Group Life in the Classroom." Norman Gronlund also published his influential book,Sociometry in the Classroom in 1959.

After the "Equality of Educational Opportunity Report" (1966), sometimes referred to as the "Coleman Report," was released, several court decisions resulted in the racial integration of many public schools. At this time the concerns of many researchers became focused upon intergroup relations in the public school classrooms. With the advent of Public Law 94-142, sometimes known as the "mainstreaming" law, similar interests in the integration of handicapped children into regular or normal classrooms became an important interest for researchers. In the 1990's the term "inclusion" is being used to describe similar processes. Thus, throughout the 1970's and continuing well into the 1990's many research activities continue to be interested in the well being and health of classroom groups and the individuals who occupy these groups.

Especially during the 1980's several books and research articles have been produced which have demonstrated much refinement in the techniques of sociometric measurement, and have also given us further insights into both individuals and group processes. Steve Asher and John Gottman's (1981) book, The Development of Children's Friendships, and the entire Summer issue of the Merrill-Palmer Quarterly (1983) offer a great wealth of information concerning theoretical and empirical information about sociometry. One important contribution in this area is Children's Peer Relations: Issues in Assessment and Intervention (Schneider, Rubin &;Ledingham, 1985).

Past research is indicating that healthy classroom climates also appear to be related and likely to enhance overall classroom academic achievement (Schmuck &;Schmuck, 1996). There are curriculum which have been developed which can improve classroom climates: for example Vacha, McDonald, Coburn &;Black's (1979) book, Improving Classroom Social Climate. Coie and Dodge (1983) have presented evidence that children's social status, developmentally across time, remains quite stable, especially children who are not accepted by their peers. This social rejection has been shown to be a strong predictor of classroom absences, later school drop-outs, and a variety of other socio-emotional problems. Two recent "meta-analyses" (Newcomb, Bukowski & Pattee, 1993; Newcomb & Bagwell, 1995) bring together a rather large body of research that summerize past findings. These children have been described as being "socially at risk." Many of these findings have lead researchers towards developing a variety of intervention techniques, directed at both groups as well as individuals. The examination of basic social skills or competencies, which appear to be strongly related to peer social attraction, have lead to some vary promising intervention approaches.

Early identification of children likely to be experiencing social rejection and peer neglect is desirable. As in the case of early identification of developmentally handicapped children - sometimes described as "children at risk" - children who are not accepted by their peers may be thought of as being "socially at risk," too. Inasmuch as the sociometric measurement techniques have been refined and become somewhat easy to administer, school districts and classroom teaching have once again become interested in sociometry.

Teachers know that the groups of children they work with are more than an aggregation of individuals. They know that the groups have form and structure: that there are patterns of sub-groups, cliques, and specific friendships. Some children are liked by their group more than others. Some are also less liked and often even rejected by their group. The patterns of friendship and rejection play an important role in determining how the group will react to learning situations, and to various types of group management techniques which teachers might wish to use.

Although teachers are aware of the obvious and dramatic aspects of group structure, the more subtle inter-personal relationships may be difficult to detect. A variety of sociometric techniques are designed to bring these relationships into view. Sociograms derived from "positive" and "negative" nomination techniques and social distance Ratings are two means by which professionals may gain some insight into these relationships. The use of these techniques are but a means to further study in most cases. Understanding group structure is but a step toward studying group dynamics, and understanding group dynamics is a means to better group management and curriculum development. Teachers should not undertake the use of these sociometric techniques unless they are prepared to check their findings by careful observations, and then utilize their increased knowledge as a means to provide better living and learning situations for children. If teachers set as a goal "improved classroom climate," then sociometric techniques are a means by which they can measure whether or not they have attained this goal.

The objectives of this document are to show how classroom teachers might sociometrically measure and diagnose the peer relations, friendships, and social status of individuals in their classrooms. A variety of sociometric techniques are presented along with some examples of how one might analyze the data once they have been obtained. In addition, some sample templates designed for duplication and reproduction are provided in the appendices so that persons who wish to use these measurements in their own classrooms may do so. Some example sociometric research is also provided in the appendices. An up-to-date bibliography of references is also provided for anyone interested in pursuing this research interest.

Lastly, I would like to acknowledge with great appreciation the help of the children, faculty and administration of the Edward Holmes McGuffey Laboratory School . Because of the necessary confidentiality and cooperation involved in longitudinal sociometric research, this facility was believed to be invaluable. The data collection techniques developed by me in this facility between 1979 and 1983 provided many insights into sociometric measurement. Up until June, 1983, this facility for research and pre-service teacher training was operated by the School of Education and Allied Professions of Miami University. At the end of the 1982-83 school year, much to my regret, Miami University decided to no longer operate the laboratory facility and thus choose to dissolve it. Doing research in the McGuffey Laboratory School proved to be a rich and beneficial opportunity for myself as well as the educational research enterprise in general. As usual, there were many more research questions left unanswered then were answered. Thus, the loss of such a research facility will forever be painfully remembered.



SOCIOGRAMS ARE?

What is A Sociogram?

A sociogram is a charting of the inter-relationships within a group. Its purpose is to discover group structure: i.e., the basic "network" of friendship patterns and sub-group organization. The relations of any ne child to the group as a whole are another type of information which can be derived from a sociogram. A sociogram's value to a teacher is in its potential for developing greater understanding of group behavior so that he/she may operate more wisely in group management and curriculum development.

Sociograms may be constructed in a variety of ways. The methods described here are ones which teachers have used and found not too difficult or time consuming.

The Data Base.

The basic material from which a sociogram is constructed is collected from group members who answer questions such as these:


These questions are examples of the "positive nomination" techniques: positive in the sense that you "like" or would "like to do..." something with these children. When children's choices are restricted to only so many nominations - in our examples above we restricted the choices to three - the technique is described as a "fixed positive nomination" technique."

Some researchers recommend the use of negative questions in order to discover interpersonal resistance. An example would be "Which three people do you like the least? This would be described as a "Fixed Negative nomination" technique. If the children were also asked to rank the individuals from "most" to "least" disliked, we would call it a "Fixed rank, Negative Nomination" technique. The study of resistance or rejection is sometimes accompanied by unfavorable emotional reaction by children. Nevertheless, negative nomination information can be quite informative to the teacher. Similar information my be obtained using Social Distance ratings which will be discussed later.


FIGURE 1 (THE DATA BASE)


NOMINATOR'S
CODE #

NOMINATOR'S
NAME

1ST
CHOICE
2ND
CHOICE
3RD
CHOICE

1.
ANN DEAL (5) NELDA (20) JUDY(12)

2.
BUD JUSTIN (15) NORRIS (19) JOHN (9)

3.
DIANE DONNA (6) LAURA (16) JUDY (12)

4.
DORIS JUDY (12) LAURA (16) JUNE (13)

5.
DEAL JERRI (14) ANN (1) JUNE (13)

6.
DICK SOL (23) JIM (11) BUD (2)

8.
GARY JUNE (13) DORIS (4) NORRIS (19)

9.
HARRY JIM (11) DICK (7) BUD (2)

10.
JOHN JUSTIN (15) BUD (2) NORRIS (19)

11.
JIM MIKE (17) JUSTIN (15) BUD (2)

12.
JUDY NELDA (20) LAURA (16) ANN (1)

13.
JUNE DORIS (4) LAURA (16) JUDY (12)

14.
JERRI DEAL (5) NELDA (20) DICK (7)

15.
JUSTIN SOL (23) DICK (7) NORRIS (19)

16.
LAURA JUDY (12) NELDA (20) DORIS (4)

17.
MIKE JIM (11) SOL (23) NORRIS (19)

18.
MILLARD SAM (22) VICTOR (24) JIM (11)
19. NORRIS JUSTIN (15) NELDA (20) JIM (11)

20.
NELDA JUDY (12) NORRIS (19) LAURA (16)

21.
PRUDENCE NELDA (20) DORIS (4) JUDY (12)

22.
SAM VICTOR (24) MIKE (17) JIM (11)

23.
SOL NORRIS (19) DICK (7) JIM (11)

24.
VICTOR SAM (22) MIKE (17) JIM (11)

How to Chart a Sociogram.

STEP 1

Using a survey sheet similar to that shown in Appendix A, responses of individual children within a group to questions such as those described above may be cut into individual strips. Note that the child who is responding to the question writes his/her name in the left most block, followed by his/her three nominations in the adjacent three blocks. Thus, the block second from the left indicates the first nomination, the second nomination is the third block from the left, and the right most block is their third nomination. For purposes of later explanation a compilation of children's FIXED RANKED POSITIVE NOMINATIONS is displayed in FIGURE 1, The Data Base. The ensuing examples all draw upon this data base.

STEP 2

If one utilizes a two-dimensional matrix of "nominators" by "nominees," such as the matrix template in Appendix B and Figure 2, they can rapidly total up the frequency of nominations (positive and negative separately) which each child received from their classroom peers. After cutting the children's response strips, either directly from each child's survey response sheet (Appendix A) or from a compilation of these responses as seen in FIGURE 1 they may be sorted into alphabetical order. Then, taking each strip, one at a time, one can note who was nominated by each individual child. Entries into the matrix consist of "+" signs for each "positive" nomination and "-" signs for each "negative" nomination. If each child nominated three other children, then the sum of each ROW should never be more than three +'s or -'s. However, the sum of +'s or -'s in each column represent the number of nominations each child received and this sum should never be more than the size of the group minus 1 (children usually are instructed NOT to chose themselves and they should not nominate the same person more than once.


FIGURE 2

STEP 3

Using the total + and - frequencies obtained from STEP 2 one can construct a bar graph similar to that shown in APPENDIX C and FIGURE 3. If one gives each child a code number ranging from 1 to the size of the group (N), they can utilize the center section of the bar graph to represent each child in the classroom. Note that if one requests both "positive" and "negative" nomination data, they can plot simultaneously both the frequency of negative nominations extending downward, and the positive nomination frequencies extending upward.

FIGURE 3

A bar graph such as this can be informative. Specific patterns of rejection and attraction may be noted.

A. POPULAR CHILDREN.

Children who receive many positive nominations and very few if any negative nominations (a high positive nomination bar and a very low negative nomination bar) are usually described as "popular" children. This would approximate the "A" portion of the Target Graph described in STEP 4.)

B. REJECTED CHILDREN

Children who receive many negative nominations and very few if any positive nominations are usually described as "unpopular" or "rejected" children. (This would approximate the "D" portion of the Target Graph described in STEP 4.).

STEP 4

Using the data compiled in STEP 3's bar graph, as well as the total +'s and -'s found at the bottom of STEP 2's matrix, one can go on to graphically display these results using the "TARGET TECHNIQUE." APPENDIX D displays the traditional "Bull's Eye Target" template. The advantage of using this relatively simply and quick display is that one can group children into categories similar to those described above in Step 3. Each portion (ring) of the Target Graph is lettered "A" through "E" and these letters correspond to the following definitions:


FIGURE 4

A. (POPULAR CHILDREN):


This is the "bull's eye" portion of the graph and is reserved for the most "popular" children. Popular is defined as children who have a frequency of "positive" nominations great than 3, the average number of nominations if children evenly distributed their choices over the entire class, and have near zero (0) "negative" nominations.

B. (LIKED MORE THAN DISLIKED):


Children who have more "positive" nominations than they have "negative" nominations go in Sector B.

C. (DISLIKED MORE THAN LIKED):


Children who have more "negative" nominations than they have "positive" nominations go in Sector C.

D. (CONTROVERSIAL CHILDREN):


A special kind of category called "controversial" is reserved for children who have nearly an equal number of "positive" and "negative" nominations and both exceed the average nominations if children evenly distributed their choices over the entire class.

E. (REJECTED CHILDREN):


Sector E is reserved for children who received only one or no positive nominations and their negative nominations exceed the average number of nominations if children evenly distributed their choices over the entire class.

F. NEGLECTED CHILDREN

Children who do not receive any "positive" nor any "negative" nominations should be placed in the F sector. These children are not even being acknowledged by their classmates. Nevertheless, this is useful information, but as one can see, using the "positive - negative" nomination technique tells us very little about NEGLECTED children. As we shall see later, the classroom social Distance Rating technique could give us some further insight into these individuals.


STEP 5

First, fold each response strip in the middle so that only the nominating children (in the left-most block) and their first choice nomination (in the adjacent box) can be seen (see FIGURE 5).


FIGURE 5

STEP 6

Next arrange the response strips in a top down descending order from those children who received the most nominations to those who received the least nominations. Inotherwords, you will be placing the folded strips in a column, one strip below the next, and this should display those children who have been chosen the most - they should have the same names or code numbers in the second column. Your Target Graph, Sector A, should be a good clue as to who should be on top. These are called "sets" and do not include children who did not receive any "first place" nominations - those occupying Sector F in the Target Graph. There are three slips in the first set including Bud (02), John (10) and Norris (19). The arrangement should look like FIGURE 6.


FIGURE 6

STEP 7

Chose one of the sets having the most slips! Place the slips on a large sheet of paper with the right hand sides of the slips - the folded edges- converging as in FIGURE 7. It does not matter how these are placed so long as all the names/numbers are together. This is an approximate configuration and may be shifted as other slips are added in later steps. It is usually wise to start the pattern with the set carrying the name of the "most chosen" girl in the center of one half of the paper and the set with the "most chosen" boy in the center of the other half. These will probably be the same children you have placed in Sector A of the Target Graph. Usually all the other slips will fall in place around these two sets.


FIGURE 7

STEP 8

The slips displayed in Figure 7 show who nominated Justin (15). The next step is to see whom Justin (15) chose. This may take some hunting. Look through the slips left in the original column until you find the one with Justin's name on the left most side of a slip. In this example Justin's choice is in the set with Sol's (22) name on the right. Place this on the diagram with Justin's name in the center (FIGURE 8). Note that when you position this slip the folded edge is away from the other slips.


FIGURE 8

STEP 9

The next step is to find whom Sol (22) chose (the slip with Sol's name on the left) and place it with the two "sol's" together (FIGURE 9).


FIGURE 9

STEP 10

Following the plan of finding the person chosen by the last in line, the hunt starts for the slip with Norris' (19) name on the left. Since no slip of this nature can be found in the column or in the sets, it means that this slip must already be charted. Examination of the layout shows this to be true (FIGURE 9): Norris chose Justin). The two "Norris'" must converge, so readjustment is necessary. Slips for Bud (2) and Norris (19) are reversed in position, and the Sol/Norris slip is moved to reath the Norris/Justin slip as in FIGURE 10.

FIGURE 10

STEP 11

When the slip with Sol's name on the left was located (Step 8) it was found to be in a set. This set now has one slip remaining (Dick --- Sol). This slip is placed with the "Sol's" converging as in FIGURE 11. Every slip in this pattern now has been matched. Check by seeing that there are no folded edges or slips which are not matched with another slip.

FIGURE 11

STEP 12

Leaving the slips of this first pattern, start diagraming another set. For example, there are two slips in the set with Jim's (11) name on the left. Jim (11) has chosen Mike (17), so there is a mutual choice (Mike --- Jim, and Jim --- Mike). This is an example of a Mutual Choice. Place these side by side with the two "Jim's" together and add the third slip (Harry --- Jim) so that the "Jim's" converge as in FIGURE 12.


FIGURE 12

STEP 13

Working through all the sets described in STEPS 4 through 12 above, it may be necessary to do some shifting of total patterns in order to configure the resulting patterns in a fairly even distribution. If. after all the slips from all the sets have been utilized, some slips may remain in the original column. Place them in the pattern where they belong: ie., where the name or code numbers on the right of the remaining slips match the names or code numbers on the left of a slip already in place.

A WORD OF CAUTION

Sometimes children choose individuals in the group who were absent at the time the basic material was collected. This means that there will be no slips with these names on the "left". Make an additional slip, writing in each such name and marking "absent" on it. Some children may misunderstand the directions and name individuals not in the group. Make additional sips for these names, marking "not in group." Place these slips on the chart adjacent to the slips of the persons who nominated there names. These cautions allow us to account for all the children's nominations.

STEP 14

Now that all slips have been placed on the chart, make a final check to see that no folded edges are left unmatched. Rearrange (by shifting all related slips) any patterns which seem to be crowding each other. If there are any cross-sex choices, note where they occur and position them on a border position adjacent to the child of the opposite sex whom they chose. Children who did not receive any first place nominations should be placed in peripheral positions. With these strategies in mind it is time to introduce some of the conventional symbols utilized in drafting out a sociometric map.




The main idea of a sociometric map is to show who chose whom. This is usually accomplished by connecting a line from one person to another: the one doing the nominating is connected to the person they nominated. This provides us with a visual graphing of the "networks" of friendship. Since children are usually asked to nominate three people for each sociometric question, each level of choice (first, second and third) is represented by a different type of line. An arrowhead at the end of the line pointing towards the person who was nominated represents the direction of the choice. An arrowhead at both ends of a line indicates a "MUTUAL" choice.

STEP 15

Using the above symbols, the next step is to transfer the obtained slip patterns to a "drawing". If you want to do a neat job and are not too sure of your drawing ability, an easy trick is to use small round and square gummed labels. Remove the slips of paper, one at a time, and place a circle, or a square, for each name or coded number slip, write the name or code number in the figure, and draw a line connecting the circles or squares with an arrow pointing to the name or number chosen. For example Norris chose Justin, so the arrow points to Justin in FIGURE 13. NOTE: REMEMBER THE PERSON WHO MADE THE CHOICE IS ALWAYS ON THE LEFT; THE PERSON CHOSEN IS ON THE RIGHT, TOWARD THE FOLDED EDGE!


FIGURE 13

Continue until all slips have been removed and the total pattern has been transferred to paper (FIGURE 14). This forms the basic pattern for the sociogram. Second and third choices are elaborations superimposed on this pattern... with some re-arranging.


FIGURE 14

STEP 16

You are now ready to plot the second choice nominations. Fold the slips so that the children's second nomination is adjacent to the name (number) of the child making the choice as in FIGURE 15.


FIGURE 15

Now you must make a decision as to whether you want to give equal significance to all nominations (first, second and third nominations). Your decision will depend on what you want to know and how you plan to use the sociogram. Thus, there are two strategies: (A) no distinction is made between the three nomination ranks, and (B) you want to symbolically distinguish between the three nomination levels. Giving equal significance makes a sociogram easier to read. Distinguishing among nominations provides greater precision. For example you may want to know whether a person chosen by three others was chosen first by three others or third in each case. Examples of sociograms NOT making any distinction between nomination levels are contained in FIGURE 16.

FIGURE 16

Sociograms distinguishing between nomination levels are contained in FIGURE 17. Both FIGURE 16 and 17 represent the same data.



FIGURE 17

A. EQUAL SIGNIFICANCE GIVEN TO ALL 3 NOMINATIONS.

With the slip folded to cover the first choice (FIGURE 15) find the basic pattern you have plotted, the name on the right, and draw a line from that name to each of the other two names on the slip. For example, draw lines with arrows from Bud (2) to Norris (19) and from Bud (2) to John (10) as in FIGURE 16. If you find that there is already a line between the names, use arrowheads at both ends to indicate mutual choices (e.g., Dick (7) and Sol (23), Norris (19) and Nelda (20), as in FIGURE 16. BEND your lines and circumvent squares and circles which get in the way of a straight line (eg., Dick (7) and Bud (2), Donna (6) to Judy (12) in Figure 16. Sometimes a minor rearrangement of the circles and squares can avoid clutter. Sometimes clutter cannot be avoided. This is especially true of negative nomination sociograms.

B. DISTINCTION IS MADE BETWEEN NOMINATION LEVELS.

FIGURE 17 displays all three nominations. Find connecting names as in "A:" above, but draw "thin" dark lines to indicate "second" nominations. Do the same for third place nominations but use a "dashed" line. Sometimes different colors are used to distinguish between first, second and third place nominations. One problem with colors is that they will not reproduce as easily on a copying machine. For this reason alone it is recommended that colors not be used. And, color costs more money!

Whether or not one uses technique "A" or "B," plotting a sociogram is usually easier if all second choices for all slips are done first, and then do the third nominations last. Of course, folding the slips so that the nominators and their third choices are visible will aid in the plotting.

One additional word on the issue of second and third choices. When children are instructed to rank their nominations (eg., "Make your first choice the person whom you like the best, and your second choice the person you 'next' like best, etc."), a weighting scheme may be applied to their nominations. This technique is sometimes referred to as "Fixed Rank" nomination technique: remember "fixed" in that the children are limited to 3 choices, and "Ranked" in order of priority. Sometimes 3 points are assigned to "first-choice" nominations, 2 points for a "second-choice," and 1 point to a "third-choice". For each individual child the frequency of nominations within each rank is multiplied by the assigned weight and these are summed to yield an index of popularity. If a child receives no nominations their score would be zero (0). Many first choices would tend to yield a high index. If the children's nominations are entered as "weights" replacing the "+" and "-" signs with their respective weights into the matrix described in STEP 2, simply summing the weighted values in each column would yield each child's weighted index of popularity within the group.

SOCIOGRAM INTERPRETATION AND TERMINOLOGY

Giving examples of sociogram terminology is one of the easiest ways of relating how to interpret a sociogram. All of the examples which follow are taken from FIGURE 14. One might note that the basic terminology which follows can be broken down into two categories, Stars, Isolates and Ghosts (A, B and C) are terms which describe individual children or INDIVIDUAL PHENOMENA, while mutual choices, chains, islands and triangles (D, E, F and G) are attributes of social interaction within a group or GROUP PHENOMENA.

  • 2. GROUP PHENOMENA


    Note the two groupings which seem to have the same construction in FIGURE 14, each with a mutual choice and someone who is just hanging on: Harry - Jim - Mike and Millard - Sam - Victor. Turn now to FIGURE 16 or 17 and note the difference between the two groups. The Millard-Sam-Victor group maintains its individuality, remaining an ISLAND, while the Harry-Jim-Mike group is much more integrated with the group as a whole and is definitely no longer an ISLAND as indicated by the many nominations received by Jim from children outside the original ISLAND configuration. Victor is the only child in the other island who receives an outside nomination. The Millard-Sam-Victor combination proves to be a definite sub-group, while in the Harry-Jim-Mike combination, we see that Mike and Jim are mutual friends and Harry is an ISOLATE, not accepted even by those whom he nominates. Looking at Harry's position in FIGURE 17, we can observe that he seems to be attempting to find acceptance in widely separated groups. This is not unusual behavior for an ISOLATE who is desperately reaching out. Mike would also be designated an ISOLATE if it were not for his being chosen by Jim. Jim is definitely a STAR. Note that Jim is also accepted by Justin's friends, Sol and Norris, but not by Justin, even though Jim chose Justin. Actually Jim is the only outsider chosen by the Millard-Sam-Victor ISLAND, all of whom chose him after themselves. As is indicated by the lines leading toward his name, Justin is the person of status in this group, accepted by influential individuals, but not necessarily accepting them. Jim may very well be exerting more functional leadership with more people than Justin.

    Looking at the Girl's side of FIGURE 14 note that Laura would seem to be an ISOLATE, but in FIGURE 16 and 17 she shows up as a star or co-leader in the group. Also, contrary to Jim among the boys, she is accepted by the status leaders (those who are chosen by influential individuals but do not return the compliment). We also have here an example of an ISLAND pair, Donna - Diane. As can be seen in FIGURE 16 or 17, both Donna and Diane make identical second and third choices (Laura and Judy). This pattern of similar choices is not unusual in such situations. There are also two isolates among this group of girls, Gary and Prudence. Jerri would be an ISOLATE if it were not that Dael had chosen her. At this age, non-acceptance by the group and choice of girls by boys is often an indication of immaturity. This is less likely to be true of Nelda, who also choose a boy, as she has a high degree of acceptance among the girls, who at this age may prize maturity, particularly physical maturity. Moreover, since Norris has chosen her in return, it may be that these two are more mature as 6th graders than their fellow students.

    These example may serve to indicate the type of analyses possible when a sociogram has been plotted. Such analyses lead to further observation and study.

    COEFFICIENT OF CLASSROOM COHESION


    Vacha et al (1979) have described group cohesion as:

    "...the attraction structure of the classroom and involves not only individual friendships but also the attractiveness of the whole group for individual students. In cohesive classrooms, students value their classmates, are involved with and care about one another, try to help one another, and are proud of their membership in the group. Student cohesiveness can either support or undermine educational goals depending on the impact of other group processes in the classroom. For example, if students share counter educational norms that limit student participation or undermine academic achievement, their cohesiveness can work against the academic goals of the schools by making those norms extremely difficult to change. If a classroom group develops norms that support academic achievement, high cohesiveness can enhance education by providing a strong 'we feeling' which promotes conformity to student norms." (p. 221

    Vacha et al (1979) suggest three patterns of classroom social relations which they believe are typical threats to classroom cohesion. They include:

    Upon sociometrically surveying a classroom through the use of the "positive" and "negative" nomination techniques, one should analyze the evidence for any serious social cleavages, in-/out-group rivalries and divisive individual competition which might threaten classroom cohesion. If these cliques are not present, then the "coefficient of cohesion" ("C") may be computed. This computation is an indicator of how strong the mutual ties are among the classroom members, and is based on the obtained number of mutual choices. Vacha et al (1979) suggest that "There is no objective criterion that can be used to determine whether or not a given coefficient of cohesion indicates the existence of a problem in any particular classroom." However, their experience in administering sociometric measures in many classrooms at the 4th through 6th grade levels provides a convenient rule of thumb. The coefficient of cohesion of 19 classes ranged from a high of 15.58 to a low of 3.83. Their median coefficient was 6.12, and the mean coefficient was 7.1. Based on their experience you may wish to consider a class as having a cohesion problem if it's coefficient of cohesion is below six or seven.

    The coefficient of cohesion can be calculated directly from sociometric data used to diagnose "positive nomination" data. All of the data necessary are contained in the sociogram (primarily as in FIGURE 16). To calculate the coefficient of cohesion, simply count the number of mutual positive choices made by all of the students, the total number of positive choices made by all of the students, and the number of students who completed the survey. The coefficient of cohesion can then be calculated using these totals according to the following formula:

    C = Mq/Up = (15*.87)/(57*.13) = 13.05/7.41 = 1.76



    Where:


    VARIATIONS ON THE POSITIVE/NEGATIVE NOMINATION TECHNIQUES


    THE RECOGNITION SCALE: Besides positive and negative nominations for friendships, some researchers ask for nominations for a variety of different behavioral characteristics or attributes. One such instrument, The Recognition Scale, has been used by Sherman and Burgess (1985). This scale is similar to a "Guess Who" nomination technique in which children nominate classmates who fit descriptions of behaviors. It was constructed by combining descriptions from The Ohio Recognition Scale (Fordyce, Yauch and Raths, 1946) and from similar research instruments by Johnson (1950), Johnson and Kirk (1950), and Baldwin (1958). This scale includes a list of students in a classroom (similar to the Social Distance Scale described next) which has a list of 20 behavioral attributes which describe student behaviors. Teachers read a paragraph describing a particular behavioral attribute that corresponds to the name on the children's response form. Students are instructed to check up to three different students who might fit the particular description which the teacher has read to them. They are told that they do not have to choose anyone for a given trait if they feel that it does not fit anyone in their room. Students' profiles are obtained by adding the total number of nominations they each receive on each attribute.

    Many of the traits in the Recognition Scale describe basic social competencies that have been shown to be strongly related to children's acceptance/rejection and general social status within the classroom. In a recent article by Sherman and Burgess (1985) a strong relationship was shown between several clusters of these traits and children's Classroom Social Distance Scores (see Table 1). Coie, Dodge and Coppotelli (1982) also presented evidence demonstrating similarly strong relationships between several basic social skill traits and both positive and negative nomination data (see Table 2). This technique can be quite helpful in identifying the reasons why particular children maintain their classroom social status. Intervention using a variety of social skills training procedures can be helpful in changing the classroom climate. Work with individual children may sometimes be required. Work with the entire group is also sometimes advisable (see Vacha, et. Al., 1979, for an entire two year curriculum focusing on enhancing 4th to 6th grade children's classroom climates.

    NON-VERBAL TECHNIQUES. Sometimes teachers in the lower elementary grades (pre-school and kindergarten through the second grade) find that their children cannot read a list of names. One solution is to take photographic portraits of the children, either using the annual year book pictures or sometimes taking fast developing polaroid photographs. Then, the children are verbally asked the traditional positive and negative nomination questions with the added instruction that they select the three photographs of children they like the most or least. This technique requires an individual administration to each child, along with a little more direct recording of the children's responses by the teacher. From here on, though, all the rest of the procedures are the same. Shelly Hymel's (1983) article dealing with sociometric measurement in the preschool environment is a good reference for those interested in this area. Hymel (1983) also discusses another technique called the "paired-comparison" measure. "Hear a child is presented, in turn, with all possible pairs of peers within the group under consideration, and for each pair is asked to state a preference for one or the other peers according to some specified interpersonal criteria (e.g., '"Which one would you most like to play with?')," (p. 24) One main disadvantage to this technique is the amount of time required for administration, especially with preschoolers!

    ANOTHER WAY OF DETERMINING SOCIAL STATUS


    Coi and Dodge (1983) have developed another way of determining children's social status within classrooms which is based on quantifiable measures obtained from positive and negative nomination data. Their formulas are useful when the most general nomination format states "What three children in this classroom do you 'like the most'" or "What three children in this room do you 'like the least'". Nominations may be treated as 'fixed nominations' (for instance fixed at a maximum of three nominations) or 'fixed-rank' measures (for instance giving a weight of 3 to first nominations, 2 to second nominations, 1 to third nominations and 0 for no nominations). Certain basic statistics such as the mean and standard deviation can be calculated for each general question: i.e., the mean and standard deviation for 'liked the most' as well as 'liked the least'. In order to use Coi and Dodge's (1983) formulas for determining social status you must first convert the children's raw frequency or weighted frequency scores to standardized z-scores. This is fairly simple to do after one has computed the means and standard deviations for each of the two general questions 'liked the most' (LM) and 'liked the least' (LL). The mean for each question is subtracted from each child's score on that question and this difference is divided by the standard deviation for that question:

    Child's score - mean
    --------------------------------- = z-score
    standard deviation

    zLM = the z-score for the 'liked most' question.

    zLL = the z-score for the 'liked least' question

    SOCIAL PREFERENCE (SP). After computing each child's zLM and zLL scores one can then go on to compute each child's Social Preference score which is as follows:

    SP = zLM - zLL

    The main idea behind this construct, Social Preference, is that the zLM score will obviously be a positive figure if zLM is greater than zLL, while if the zLM is less than zLL, the result will be a negative number. Negative numbers exist at one end of the continuum indicating very little preference, while positive scores indicate a strong preference. Scores in the middle, that is close to zero (0) indicate average social preference.

    SOCIAL IMPACT (SI). Social Impact is the idea that whether or not children are liked by the group as a whole, or disliked by the group, the group is, nevertheless responsive to them. Therefore, both the zLM as well as the zLL scores are an important consideration. Thus the formula for Social Impact is as follows:

    SI = zLM + zLL


    Given these four computations [1) zLM, 2) zLL, 3) SP and 4) SI] one can then apply Boolian Logic to determine the five specific categories of social status:


    One distinct advantage of standardizing children's scores - i.e., transforming their raw positive or negative nomination scores into "z-scores" - is that these z-scores make possible normative comparisons of individual in different environments. Obviously one factor effecting the children's raw scores is the size of the classroom group with which the child is being measured: eg., the difference in frequency of nominations in a large group may be substantially different than the frequency of nominations they receive in a different and "smaller" group. Sometimes teachers and wish to know the stability or similarity of a child's social status in different classrooms, or even over time (eg., do kids change or maintain their social status as they grow older?). The standardized z-scores provide an appropriate way of answering questions such as these because these standardized scores reflect normative standing in a group regardless of the constituency or size of the group.

    SOCIAL DISTANCE: A SOCIOMETRIC RATING SYSTEM

    Asher and Hymel (1981), Kane and Lawler (1978), as well as Miller and Gentry (1980) have discussed several different techniques for measuring peer attraction. An adaptation of a sociometric rating scale developed by the Horace Mann-Lincoln Institute of School Experimentation (Bureau of Publications, 1947) entitled the Classroom Social Distance Scale, is one such measure. The scale is modeled after Bogardus (1928) sociologically oriented strategy and allows each child within any particular classroom to give, as well as receive from every child in the classroom, a rating on a 1 to 5 continuum. The rating continuum is as follows:


    An example of the rating form appears in FIGURE 18 (this would be Ann's responses as in our example classroom displayed throughout FIGURES 1 - 17), and a blank template for reproduction/duplication appears in APPENDIX E AND F. Students are given a survey-matrix (FIGURE 18) in which the columns consisted of an alpha/vertical list of the children in their room, and the rows are labeled in the left margin with the 5-point rating continuum. Children are asked to indicate the statement which most nearly defines their feelings about each person. To identify who is doing the rating, each child is asked to circle their name. They are asked to "...put an astrix, '*' in the row which describes how you think most people would rate you." This is sometimes called the Personal Social Distance Rating. Great discrepancies between the Personal Social Distance Rating and the mean Classroom Social Distance ratings may indicate that a child is "out-of-touch" with their "Social Reality". As can be seen in FIGURE 19, a tally sheet made up from the master survey matrix presented in FIGURE 18, each child's mean social distance score can be computed by multiplying the appropriate weight times the frequency with which a child was checked off in each of the categories, and dividing by the number of raters (the size of the classroom minus one - the child who is being rated). Theoretically, the mean social distance scores, a continuous measure, can range from 1 to 5 and relatively low scores (1) would indicate less social distance, while relatively high scores (5) would indicate greater social distance. This social distance measure can then be analyzed contingent upon various attributes of both the raters' and ratees', such as their gender and age. Sherman (1984a, 1984b, 1984c, 1985), for example, has shown the importance of considering gender and age as potential moderating variables influencing the ratings which children both give and receive.

    The Classroom Social Distance Scale has many advantages over the positive or negative nomination technique discussed earlier. It's primary advantage is that every child within any given classroom contributes to the score each individual receives. When either the positive or negative nomination technique is used we see the "network" of friendships, but if no one nominates a child that is all we know about how others feel about him/her. A rating system is, then, much more precise.

    SOCIOMETRIC RANKING. Another form of sociometric measure, somewhat similar to the sociometric "rating" strategy is called the "Sociometric Ranking." Here, once again, the children are given a list of names of their classroom peers. They are asked to "rank" the names from lowest to highest on the basis of some interpersonal criteria (eg., Rank the children from highest to lowest on the basis of who you would most prefer to play with, where a low ranking could indicate the lest preferred and a high ranking the most preferred.) The rankings which children receive from their peers are summed and divided by the number of children who did the rankings (the classroom size minus 1) to yield an "average rank score."

    OBSERVATIONAL SCHEMES. A sensitive teacher/researcher can also obtain a sense of children's popularity and social status simply by observing who interacts with who. This is done in free-play situations where the children have the freedom to interact with their peers: eg., at lunch time noting who sits with who; during recess noting who plays with whom. Sometimes the validity of social interaction is evaluated noting whether the interactions are positive approaches or negative ones. Cavallaro and Porter's (1980) research on preschool "at-risk" and normal children's peer preferences is an excellent example of naturalistic observational sociometric analysis (See Appendix G, Research Example No. 7). Observational analyses are sometimes used to confirm children's "paper-and-pencil" sociometric nominations and ratings.

    APPENDICES



    Appendix A: Sociometric Nomination Form
    Appendix B: Positive/Negative Nomination Tally Matrix
    Appendix C: Positive/Negative Nomination Bar Graph Form
    Appendix D: Positive/Negative Nomination Target Graph Form
    Appendix E: Social Distance Form
    Appendix F: Social Distance Tally Form
    Appendix G: Recognition Scale

    REFERENCES

    This WWW site has been constructed by Lawrence W. Sherman. I wish to acknowledge the support of the Center for Human Development, Learning and Teaching AND the Department of Educational Psychology. Please send any comments and suggestions about this home page to Lawrence W. Sherman.

    This Document was last modified on Tuesday, January, 13 1998 at 15:15:55.
    This Document has been accessed 35 TIMES.