CO NNECTIONS 19(1):39-42

©1996 INSNA

Some Antecedents of
Social N etw ork
Analysis 1

Linton C. Freem an
University of California, Irvine

Writers often su ggest that m od ern social netw ork analysis began w ith the p u blication in
1934 of Jacob L. Moreno's pioneering book on sociom etry, W ho Shall Survive? (Alba, 1982;
Freem an, 1989, p.17; Wasserm an and Fau st, 1994, p . 12; Degenne and Forsé, 1994, p. 31).2 The
p u rp ose of the present note is to show that, before 1934 w hen his book w as pu blished , there
w ere antecend ents for both Moreno’s ow n id eas and for a great m any of the m ore recent id eas
of social netw ork analysis.

My aim here is not to belittle the trem end ou s im p act of Moreno's w ork on ou r field , bu t
only to su ggest that several others—p eople w orking prim arily in ed u cational and
d evelop m ental p sychology—d eserve cred it for the origination of m any of the id eas and
m ethod s w e now use in netw ork analysis. I w ill review six of these pre-1934 innovative
contribu tions here:

1. Almack, J. C. 1922. The influence of intelligence on the selection of associates. School

and Societ y , 16:529-530.

In 1922—even before Moreno started thinking abou t his m ethod —John Alm ack, a p rofessor
at Stanford , pu blished a pap er that anticip ated the d evelopm ent of the sociometric instru m ent.
Alm ack asked child ren, grad es 4 to 7, in a California school to answ er a series of qu estions
abou t those w ith w hom they w ou ld like to w ork and those w ith w hom they w ou ld like to play.
One qu estion, for exam p le, w as, "If you had a p arty, w hich boy from you r class w ould you
invite?"Alm ack tabu lated chooser-chosen pairs and then ran correlations betw een the I.Q.’s of
p airs and exam ined the still-cu rrent hyp othesis that choices are hom op hilou s.

1
I am g r atefu l to D ev o n Brew e r w h o first ca lle d m y a tte n tion to th e stru ctu ra list lin e of
r esea rch in d ev elo p m en ta l an d ed u ca tion a l p sy ch olog y .

2
W a sser m a n an d Fa u st also in d ica ted th a t M or en o in tr od u ced th e essen tial id ea s b eh in d
so cio m etr y p u b licly a y ea r ea r lier a t a m ed ica l co n v en tio n . A n d D eg e n n e a n d Fo r sé p o in ted o u t th a t
in th e p r efa ce to th e Fr en ch la n g u a g e ed ition of h is b ook (Fon dem en ts de la sociom étrie, 1954), M or en o
w r o te th a t h e h a d b eg u n w o r k on th e sociom etr ic p er sp ectiv e in 1923.
40 S O M E A N TEC ED EN TS O F ... / Freem an

2. Wellman, B. 1926. The school child’s choice of companions. Journal of Educat ional
Research, 14:126-132.

Like Alm ack, Beth Wellm an focu sed on hom op hilou s choices am ong p airs of ind ivid u als.
Bu t w hile Alm ack d ep end ed on rep orts of choices, Wellman record ed those pairs of ind ivid u als
w hom she had observed as being together frequ ently. Thu s her ap p roach anticipated the kind
of observational techniqu es later used by Roethlisberger and Dickson (1939) and by Davis,
Gard ner and Gard ner (1941).

She stu d ied 63 boys and 50 girls w ho w ere enrolled in ju nior high in the Lincoln School, ru n
by Teachers College, Colu m bia University. Over a period of five m onths, she w ent to d ifferent
p laces arou nd the school and record ed w ho she cou ld observe interacting w ith w hom d u ring
p eriod s of free activity. She also collected trait d ata; for each child she record ed height, school
grad es, I.Q., perform ance on a test of p hysical coord ination and p osition on a teacher’s rating
scale of introversion-extroversion. Wellm an stu d ied hom op hily w ith respect to all of these
traits.

3. Chevaleva-Janovskaja, E. 1927. Groupements spontanés d’enfants à l’age préscolaire.

Archiv es de Psy chologie, 20:219-223.

Eu génie Chevaleva-Janovskaja d esigned the first large scale observation-based stu d y of

grou p stru ctu re. She d evelop ed a p rogram in w hich preschool teachers in Od essa w ere taught
how system atically to observe child ren’s tend encies to get together to interact. Teachers w ere
p rovid ed w ith a list of 19 item s that specified w hat and how to observe.

The item s w ere d esigned to gu id e observation and to stand ard ize w hat w as to be record ed
and how . They w ere d istribu ted to teachers and d ata w ere collected on 888 grou p s involving
276 child ren. Chevaleva-Janovskaja stud ied th e im p act of age on group form ation, and the
extent to w hich the grou p s w ere hom ogeneou s w ith resp ect to both age and sex.

4. Bott, H. 1928. Observation of play activities in a nursery school. Genet ic Psy chology
Monographs, 4:44-88.

H elen Bott’s w ork w as the precu rsor for a great d eal of contem p orary w ork in social
netw orks. She stu d ied child ren in a p reschool attached to the University of Toronto. She began
as an ethnograp her m ight and sou ght to uncover any form s of behavior that recu rred regu larly
am ong the child ren. She uncovered five su ch form s: (1) talked to another, (2) interfered w ith
another, (3) w atched another, (4) im itated another, and (5) coop erated w ith another.

Bott set about the task of system atic observation of these behavioral form s. Each d ay one
child w as d esignated as “focal” and all observations w ere centered on that child and his or her
p artners. In this app roach Bott anticip ated Sm ith (1931) w ho w as the earliest u ser of focal
sam p ling found by Altm an (1974) in her exhau stive review of observational sam p ling m ethod s.
In any case, Bott tallied every instance of each form of behavior, along w ith inform ation on
w hich other child w as the target of that behavior.
 S O M E A N TEC ED EN TS O F ... / Freem an 41

As Freem an and Wellm an (1995) rep orted earlier, H elen Bott’s ap p roach anticip ated that
of Forsyth and Katz (1946) by organizing d ata into m atrices. And it an ticip ated the w ork of
Moreno (1934) by d iscu ssing the results in term s of kind s of interp ersonal linkages.

5. Hubbard, R. M. 1929. A method of studying spontaneous group formation. In Some New

Techniques for St udy ing Social Behav ior. D orothy Sw aine Thomas, ed. Pp. 76-85. N ew
York: Teachers College, Columbia University, Child D evelopment Monographs.

Ru th H u bbard p ion eered in th e stu d y of tech n iqu es for observin g in teraction . She
system atically exam ined inter-observer reliability in record ing p atterns of w ho interacted w ith
w hom am ong a collection of 18 preschool child ren.

6. Hagman, E. P. 1933. The companionships of preschool children. Univ ersit y of Iow a

St udies in Child W elfare. 7:10-69.

Elizabeth H agm an anticip ated the Bernard -Killw orth-Sailer (Killw orth and Bernard , 1976,
1979; Bernard and Killw orth, 1979; Bernard , Killw orth and Sailer, 1980, 1981, 1982) ap p roach
to the stu d y of inform ant accu racy by m ore than 40 years. Sh e first observed interaction
frequ encies—w ho p layed w ith w hom —repeated ly over a school term . Then she interview ed
her subjects and asked them to recall their p laym ates of tod ay, those w ith w hom they played
yesterd ay and those w ith w hom they p layed at the beginning of the school term .

These six stu d ies p rovid ed m od els for a great d eal of su bsequ ent research in ed u cational
and d evelop m ental psychology. This trad ition of research into child ren’s netw orks is
m aintained to this d ay (Lid d ell and Kru ger, 1989; N abu zoka and Sm ith, 1993; N ew com b and
Bagw ell, 1995). And at least since the 1940's, psychologists stu d ying these netw orks have
recognized , and cited , related research in sociom etry. Bu t those of u s w ho com e from the
sociometric trad ition seem to be pretty m u ch u naw are of this parallel line of w ork; at least w e
seld om cite psychologists w orking in th is trad ition. The unfortu nate consequ ence of this
ignorance is that w e have had to re-invent m any of the id eas and tools that had alread y been
introd u ced and ad op ted in d evelop m ental and ed u cational psychology.

References

Alba, R. D. 1982. Taking stock of netw ork analysis: a d ecad e’s resu lts. Research in the Sociology
of Organizations 1:39-74.
Alm ack, J. C. 1922. The influ ence of intelligence on the selection of associates. School and Society,
16:529-530. Altm ann, J. 1974. Observational stud y of behavior: Sampling method s. Behaviour
49:227-267.
Bernard , H . R. and P. D. Killw orth. 1979. Inform ant accuracy in social netw ork d ata II. Human
Communication Research 4:3-18.
Bernard , H . R., P. D. Killw orth, and L. Sailer. 1980. Inform ant accu racy in social netw ork
research IV: A com p arison of cliqu e-level stru ctu re in behavioral and cognitive d ata. Social
N etworks 2:191-218.
42 S O M E A N TEC ED EN TS O F ... / Freem an

Bernard , H . R., P. D. Killw orth and L. Sailer. 1981. A review of inform ant accu racy in social
netw ork d ata. In M odelle für A usbreitungsprozesse in Socialen Strukturen. H . J. H um m ell and
W. Sod eur, ed s. Pp . 153-186. Du isberg, FRG: Sozialw issenschaftlichen Koop erative.
Bernard , H . R., P.D. Killw orth and L. Sailer. 1982. Inform ant accu racy in social-netw ork d ata
V: An exp erim ental attem p t to pred ict actu al com m u nication from recall d ata. Social Science
Research 11:30-66.
Bott, H . 1928. Observation of play activities in a nu rsery school. Genetic Psychology M onographs,
4:44-88.
Chevaleva-Janovskaja, E. 1927. Grou p em ents sp ontanés d ’enfants à l’age préscolaire. A rchives
de Psychologie, 20:219-223.
Davis, A., B. B. Gard ner and M. R. Gard ner. 1941. Deep South. Chicago: The University of
Chicago Press.
Degenne, A. and M. Forsé. 1994. Les Réseaux Sociaux. Paris: Arm and Colin.
Forsyth, E. and L. Katz. 1946. A m atrix ap p roach to the analysis of sociom etric d ata. Sociometry
9:340-347.
Freem an, L. C. 1989. N etw ork rep resentations. In Research M ethods in Social N etwork A nalysis.
L. C. Freem an, D. R. White and A. K. Rom ney, ed s. Pp . 11-40. Fairfax, VA: George Mason
University Press.
Freem an, L. C. and B. Wellm an. 1995. A note on the ancestral Toronto hom e of social netw ork
analysis. Connections 18: 15-19.
H agm an, E. P. 1933. The com p an ion sh ip s of preschool child ren. University of Iowa Studies in
Child W elfare. 7:10-69.
H u bbard , R. M. 1929. A m ethod of stu d ying sp on taneou s grou p form ation. In Some N ew
Techniques for Studying Social Behavior. Dorothy Sw aine Thom as, ed . Pp. 76-85. N ew York:
Teachers College, Colu m bia University, Child Develop m ent Monograp hs.
Killw orth, P. D. and H . R. Bernard . 1976. Inform ant accuracy in social netw ork d ata. Human
Organization 35:269-286.
Killw orth, P. D. and H . R. Bernard . 1979. Inform ant accu racy in social netw ork d ata III: A
com p arison of triad ic stru ctu res in behavioral and cognitive d ata. Social N etworks 2:19-46.
Lid d ell, C. And P. Kru ger. 1989. Activity an d social behaviou r in a crow d ed Sou th African
tow nship nu rsery—a follow -u p stu d y on the effects of crow d ing at hom e. M errill-Parmer
Quarterly-Journal of Developmental Psychology 35:209-226.
Moreno, J. L. 1934. W ho Shall Survive? Washington, DC: N ervou s and Men tal Disease
Publishing Com p any.
N abu zoka, D. and P. K. Sm ith. 1993. Sociom etric status and social behavior of child ren w ith
and w ithou t learning d ifficulties. Journal of Child Psychology & Psychiatry & A llied Disciplines
34:1435-1448.
N ew com b, A. F. and C. L. Bagw ell. 1995. Child ren’s friend ship relations—a m eta-an alytic
review . Psychological Bulletin 117:306-347.
Roethlisberger, F. J. and W. J. Dickson. 1939. M anagement and the W orker. Cam brid ge, MA:
H arvard University Press.
Sm ith, M. 1931. A Study of the Unsupervised Behavior of a Group of Institutionalized Children.
N ashville: Marshall & Bru ce Co.
Wasserm an, S. and K. Fau st. 1994. Social N etwork A nalysis: M ethods and A pplications. Cam brid ge:
Cam brid ge University Press.
Wellm an, B. 1926. The school child ’s choice of com p anions. Journal of Educational Research,
14:126-132.