F.N EX1- LORn TORY STUDY OF CREli. TIVITY - IN - T~CHING STUDIES I~J FIFTH GfiliDE SOCIAL A F DELle SCHOOL SYSTEM ID 499 by JEANNETTE GhLL 1-I.DVISER - DR. RUTH HOCHSTETLER BALL STATE UNIVERSITY MUNCIE, INDInN.n l-'lliY 6, 1968 /' - The author wishes to thank Dr. Ruth Hochstetler and Dr. Fay Clardy, Jr. for their aid in the compilation and the evaluation of data in the following study. C~&~ter I 11 III ~age Introduction sod nevi8w of ~ollec ti,)n of t~e ~roLleffi. nesearct ;:.nd iVa lua liou ;:)uillr:..ctry ~nd t1-~~endix biLllographi . .. 1 o of Da ta . ............. IV v nel~teQ St~tement • . . • 1) • • 1S Conclusions . 25 . • . 33 • • j'l CHbFTER I Introduction and Statement of the Problem Creative teaching has as its task the guidance and the release of creative fotential. n teacher is cre8tive in the sense ttat he aevelors an etmosrhere that is creativity of others an ideas are accerted. at~osrhere in which every child's It is not assumed, ho~ever, thet teachers can be divided into two groufs ffiust also carefully deter~ine ~ferate for a time in c~n ~ the creative ond Nor is it assumed that any cne te3cher will teach creatively 911 the time. teacher to the In 8ther words, the crertive teacher fGnctions as a catalyst. l the uncreative. favora~le 8 It is felt t~9t s teacher the roints at which he routine way.2 be 0escribed as valuing ~ust The more crestive diverge~t thinkine m0re th2n copvergent thinking and an open-ended system more than a closed syste~.3 The Ioint has been made that the more crestive teccher also hDs a sense of security in uncertainty and 8 tolerance for amt'igulty.4 FinGlly, a cre~tive teach~r has a subjective attitude toward the subject matter and is oriented to movement from the known to the effifhasizing the form2tion of new ideas.5 un~nown, thus 2 The importance af determining the extent to which creative teaching Cbn be described rests on the value of creativity itself. This value is reflected by an abundant consensus on the need for creativity in education. This consensus may be due to the fact that society is facing new needs. 6 In the attempt to face new needs comes a warning from an educational newsletter: In today's world cre~tivity is not just a nice thing to have. It is a grave necessity. Imitating the past is not good enough; only the creativ8 society will survive.7 Emphasis on creativity as in~ovation Gnd invention in our schools then is necessary because "we c9nrwt predict what a person will need tc know in the 21st century as ~ell as in the next third of this century,,8 and because it is recognized that "our very survival defends upon the quality of the cre2.tive imagination of our next generation."9 Recent recognition of the proposed need for creativity in education is illustrated by the contrast of past Bnd present attitudes loward creativity and its encauraeement. In the past, the general impression has been that the successful development of creative ability should concern itself with persons of unusual artistic, musical, or literary endawmanto In the past, creativity has also teen regarded as a spontaneous act with no consciously directed means for its development. However, cre~tivity is now regarded 33 a goal of education. le In the categcry of recent developments ~re s~ecific admonishments as to the present state of education 3 since the recognition of the need for cre&tivlty. ~dmonishment Cne such begins -- We need to take a fresh, inventive, crestive look into the curricula of schools and colleges. It is clear that they sre geared to lIinput,·! but not to "out;,utrl. ;,~e still emph",size .Ithe dnplication of knowledge," but do Ilttle vdth creative explicatio!1., im:[:licaUon, 2nd ap~lication. We need more learning for doing. 11 This study is concerned with a possible discrepancy between fact and opinion in the area of creative teaching. Opinion, for the pllrpose of thls study, :i.s rerresented by the educational philosophies of many schoal systems which extol the virtues of cre£tivity 3nd stress the importance of creativity in F3ct is approached through an teachin~. attempt to measure the particular system. c~eutive Fsct, practices of teachers in asks rephr~sed, & "To whet extent do elerrentG:ry teacr.ers atternrt to teach creatively:'· This q ues t ion, then, forrrs the core 5 r)f the :r:robler; of th i s The problem itself required an sization of & irn~ediste definition of cre3tivity. can be defined in IT20Y either a product, rersooality, ~rocess, condition. 12 tucy . synthe- Cre0tivitv. . , which ways, is usually defined in terms of ~r envi~on~ental For the :r:urrcses of this study, however, emphasis is pl&ced on crec:tiv:ity as environmental cond1tion. ~hether approaches which are conducive to ~r 2. process and as 8.n not a teacher employs creztivi~y Dnd whether or not she establishes a climEte or environment whjch aids this process are points Jf e~rh2sis for this study. Neming the point of view taken is, however, not particularly definitive. In order to be more definite, a number of basic assumptions were compiled: 1. It is assumed that creativity deals with newness in the sense of either a person~l rediscovery of what has already been discovered Q!: the }:roduction of something entirely new.13 This implies both the transmission and the transformation of kno't/ledge .14· - 2. Such concepts as curiosity, imagination, discovery, innovation, and invention are indicative of creativjty, but cannot be equated with it.15 Creativity enco~passes them all; it is not li~ited to a single function. 16 3. Creetivity can not be copied. 17 4. Creativity refers to what is going on within the individual - his own way of thinkin~, feeling, and doing. As he devises ways of expressing ideas or feelings which to his knowledge have originated ~ithin his own thinking and are not forced by the environment, he is develcping through the process of creativity. The creative process then is Q assumed to be esse~tial to self-realization. 1u 5. It is assumed that every individu&l possesses the ability to be cre~tive. This ability, however, varies with individuals in regard to strength and/or area in ~hich they are ca~able of expressing it.lS 6. It is also assumed that creativity can not be taught, but can be released snd guided. No one ferson can tell another person how to be c:r"?E. ti ve becb use the es sence lies 't none's own Jnsjght, underst&ndings, ~nd feelings.,2Q 7. The purpose of cre&tive education, it is assumed then, is to guide creative processes which must be guid.ed (encouraged, nurtllred, allowed) .21 5 Li~it~ticns The st~dy has certain of the Study limit~tions. The decision was made, fjrst of all, not to attemrt to measure evidence of creative teaching in a]l aress of the curriculum. Such a task would be an imfossible one to perform with any degree of accuracy or validity. The most feasible apfroach, it was felt, would be to concentrate on obtaining a measure for a number of teachers in one sfecific area of the curriculum. The area decided upon was social studies. The choice \lias largely due to the nature of social studies, since it is "an integration 01 subject matter and conce.r:t formulation which calls for a bE>lance and synthesis betwel2n social and academic learning and between the immediate and the remote."22 Such a n a ree. should frovide a tr.f Ie oPI-iortuni ty for' creB ti ve teaching. 23 The desire to heve subjects as alike as possible without reference to individual differences in attitude and personality led to the choice of teachers in a specific grade level in a public school system. including children's res~onses ~ith the intention of to questions concerning practices in the classroom, grade level five was chosen. Fifth grade children were thought to be capable of unders tanding and answering any q ues tion i.neluded j n the interview, if each question were carefully analyzed as to .ocabulary level and possible ambiguity. chosen for the sarre re~son, Grade six could hsve been but was not, due to a ;reference for and interest in the lower erade by the auttor, who was also responsible for interviewing. CHuFTER II Review of Relcted Research Over the past few ye5rs there have been many studies of creativity. All of these have been concerned with the nature of the concept of creativity, the identification of creative individuals, or some specific classroom applications of creativity. The data reported reflect a s~ecific instructiona 1 setting, or the findings are of a theoretica 1 rather than a practical nature. Few of these publications report practices relative to creativity which are representative of teaching in elementary schools. There are, however, seven studies reported in the past ten years which appear to be pertinent to the determination of creative teaching. Of these seven stUdies three were found to hove a direct relationship to the problem in this study. The most recent of the three, itA Survey of Eeliefs and Practices Relative to Creativity," by William C. :""olf ,Jr. was published in March, 1964 in the Journal of Educational Research. The purpose of this study was, first of all, to find out how classroom teachers viewed the concept of creativity. Wolf's next concern was that of making a survey of learning experiences employed by teachers under the guise of creativity. 7 This study did not include the categorization of classroom teachers as creative, less creative, or uncreative, nor was the study interested in a particular area of the curriculum. In a six state area, teachers were asked by a mailed questionnaire, "what do you ttink 'creativity' is?" and "w'ihat do you do in your classroom which you consider to be creativeri". study: ficcording to the data collected in the almost half of the responding teachers conceived of crea ti vi ty as a persona 1 characteristic; one-fourth rela ted it to a product or end; while one-fifth identified the concept as a process. The data, ho~ever, crepancy in tbe teachers' statements. revealed a dis- Manifestations of creativity in the classroom did not seem consistent with the teachers' view of creativity. Inconsistencies of this nature were blamed on vaguely conceived efforts to make provision for "a nebulous concept ll in the school curriculum. The data also reported a widespread acceptance of creativity which Wolfe regarded as remarkable because the relationship between creativity and elementary education is so very recent. Froffi the data, Wolfe constructed a response pattern for the mythical "typical teacher". This teacher - 1) conceived of crea ti vi ty as e. persona 1 cha rac teris tic, 2) described creative learning experiences which pervade the entire elementary scheol curriculum, and these experiences to ex~edite instruction in the classroom. 3) employed specifjc learning and/or Wolfe concluded that: 8 Creative utilized teachers relstive learning experiences are extensively in today's elementary school; many adhere to similar beliefs and practices t04-creativity consistently in the classroorr,"~ The next oldest of the three studies, "... Guide to trinciples of Creutivity-in-Teaching with Suggestions for Use in Elementery Social Studies" by Edwc:.rd F. stone was completed in 1961. Stone compiled, from a review of literature, fifteen principles of creativity in teaching. Using these principles he built into his stJdy seven illustrations of creative exreriences in social studies which he obtained from the puLlic schools of Great Neck, New York. The seven teachers, K-6, who provided the illustrations were evaluated prior to the study creative peDple. 3S professional and These were judged according to criteria established by the investigation and supported by a jury of school personnel. Stone concluded by means of the illus- trations that one could show a valid relationship to the general principles through their successful application in actual clbssroom situations. The degree of relationship, however, was not mentioned. Stone then wrote a guide for elementary teachers which he hoped would stimulate teachers to work more creatively in the classroom. The guide also included: • • .8 sumc~tion of the theory of the creative process translated into a fur.ctionally suggestive record of certain experiences that have been offered children with successful results. 2 5 The purr-ose of the third study is apparent from its Superi0r in Dem8nstr:ting Cre~tjveness in Teachi nf l'. 1j)7 issue of tho. Journ:?l of Educations] SesearC'h. conclusions have teen thc most B:nd found stu~ies. P':'nn's of the threc c~rnrr~hensive closs relcti:nship between 9 The hi~h mean scores jn cr9stiveness and general superi')rity in te?cbjnq. He concluded that creativeness is essential ss fector to supe~ior teaching s~c('ess ~nd ~ contributing that it is rro- portiG~~lly lacking with teachers of inferjor 8biljty.26 Bond 21so c~me to the conclusion that: It is highly suggestive th~t teachers lacking strength in rl&nnlng and teaching creatively take refuge in working with supplies snd equip~ent and in testing pu[ils cn whst they -~esum~blv hpve lprrno~ c.-!. _ .... - )n .. . . t~eir L J.",,v~w~... d~/· J:~";' "" -,,-,,,, '" -'" . ~ .j". '-" The limited findjngs of 8 fourth study, one by Cnrol F. Mershsll, h,:s m":!'je it riifficult t:: eql1Ete the stt1 dy, !'Classroom Ferce:r,:ti':Ds of Hif!hly Cre,3tive ~e=chers Howev~r, it is of significance to the i' ::.--es Gnt s tlJdy b9c? use :)f the re 13 ti oDs':.i res;onses of hi~h1y found thEt crc~tive hi~hly guided il N(~ 1t!E.S f0und cetween t~acters. crentive teachers tended to Ilfo11:J1,.;s directions," attec.I-t r and less creative sugg8st~?d • Less Cre::::tive in the E]e;rrentc::.--y 8('\;.ocl,'1 with the thr,ee studtes glready described. M~::.--sh911 ::~.:1d r.'Jade "sits jn seat", 2nd "te::.:.C'h2r in tLe study, hov:ever, to exr:Ltin how either the "hie;hly cre<::tive" O~ 1'1ess c-restjve" 10 teachers were chosen or categorized. 28 The three remaining studies contain findiQgs which msy be regarded crea ti vely. l~. &S pertinent considerations for teaching Studies by three authors, !,'rank E. May, Janet Gilbert, and DonE.ld J. Eadley, produced findings on the relationship between intelligence and creativity while they were investieating relbted problems. The data used in Gilbert's study, suprorted the previous findings of Getzels, and Torrance who found that intelligence tests Jac~son, were not satisfactory instruments for identifying creative stUdents. Gilbert also hypothesized that on standardized measures of achievement, the creative student w~uld do as well or better than less creative studeGts, but that he would not fbir as well on teacher marks. 2 9 was supp,orted. This hypothesis fI.ay's findings from a study of the crestive thinking of seventh graaers agreed with those of Gilbert • .hay found tha t wi th popul,,- tions thE. t are fairly homogeneous with respect to intelligence, I.Q. scores are probably poor predictors of creative-thinking ability. between I.~. Correlations scores snd creative thinking scores in the study ranged between .08 and .30 with most of the correlations approaching zero. 30 Hadley, in a study of the relationship between creativity and anxiety, discovered additional support for the findings of both Hay and Gilbert. In his study, hadley found that a group of children who are low in anxiety and high in intelligence, were not similarly high in creativity. According to the data those with very low 11 anxiety exhibit a moderate amount of creativity; those with a slightly higher amount of anxiety exhibit the highest amount of creativity; and those with increasingly higher anxiety exhibit increasingly less creativity. Hadley also found strong evidence that: matter how much anxiety a child may feel, the introduction of some common behaviors of teachers which research indicates induce anxiety will curtail creative production.3 l ~o A major accomplishment of publications on creativity has been the labeling of sources of discouragement to creativity. is said to One such source of discouragement to creativity ~e the school itself. Data gathered by Spindler, Getzels and Jackson, Torrance, Colem&n and Henry suggest that the stUdent is being dominated by the school and instead of having his creativity nourished, his thinking, feeling and reacting are restricted. In other words, he is being stifled. 32 Two factors encountered in the school are said to be responsible for the stifling. is the work-play dichotomy. The first factor Day states that teachers suppose that there should be no playing around in work and so they do not give children many opportunities to learn creatively because lIchildren enjoy creative experiences and their I-leasure makes teachers uneasy. "33 The second factor is that of sanctions against questioning and exploration. According to the data gathered, teachers do recognize that students need to ask questions and make inquiries about their environment, however, children who do so are lIoften 12 brutally squelched" in order to "put the curious child in his place. 1I34 ticcording to Shumsky, it is the intent of ffiOSt teachers to "present a stereotyped, sugar-coated portrai t and to avoid struggle and controversy. "35 Shumsky indicates tb3t there &.re three types of teachers - the reretitive teacher, the main-idea teacher, and the creative teacher. 36 Niel poses this question, "Do we really have to stand by and watch the training of mediocrity'i"3? 8 itself. second source of discouragement is said to be society, A major obstacle is society's success orientation. The point is made that individuals are prepared by military and civilian education only for success, not for coping with frustration and failure. The inhibiting effect of this tendency "is seen again anQ again in the individual testing of chi ldren on ta sks of crea ti ve thinking. 1138 .tl.nother factor, although still dealing with success, has to do with the current pressures of society on schools to maintain high academic standards. barron indicates that the response by school people to this pressure h~s taken the form of increased pressures on children for rote memorization of facts. 3S hnother factor within society which exerts pressure on the student is his peer group. Feer groups undoubtedly value conformity, and equate divergency with abnormality. In doing so, they exert relentless pressure on one another to rid themselves of any divergent characteristics. 40 one of the most significant personal characteristics of Yet 13 especially creative children seems to be the courage to step beyond the established bounds. Thus they have to fight the peer group and often do so without teacher sup[ort. 41 I t would seem, then, the t the tea cher who a t tempts to teach creatively must do battle with society in child's peer grou~, ~eneral, the and in some cases the school, itself, and the teachers therein. In the opposite vein, points of encouragement have also teen shown by research. Lists have been compiled describing what iliUSt te done by a teacher to increase the crea ti vi ty in his teaching. These are not s tand-pb t formula s, nor are they reco;r,mend 2. tions for the use of any certain materials. Particular materials cannot be recommended because it has been concluded that too little is yet known about the effects of various instructional media on creativity.42 The creative teacher appr0aches ~nowledge with the question me as a person r( 1143 "'what does the subject matt.er meCin to Taylor summarizes the present state of affairs by saying: lt may be more effective at the present stage to develop materials designed to remove hindrances and to untrain for nO~4creativity than to directly train for creativity. In addition, Torrance has emphasized the importance of learning how to rewCird creative behavior if children are to think creatively. 45 hnd, according to Dale, the ultimate aim of education should be to try to make stUdents alike enough so that they can communicate with each other, yet different enaUEh so th:t they ~ill h~vc something worth co~~unic3tin~.~6 The issue, then, has been eVG1usted. h3ve been identified . .> rhil=so[hy of need Su~~estions exreriences. enc~urape have been ih?t ma~c to imrlement or s92~inEly remfins to be the thoughtflll rursuit ~f The bErriers t~s rr~duce ~~d belief in the rhilcs0;hy of need ~lready formul&te~. creative cc~o~rlished the rrccess instill, most certainly to test, teen to ~nd ultimately to is to ~erhaps u~h01d fcrmul&tsd. CHAfTER III Collection snd Bvaluation of Data lhe processes used in collecting Gnd evaluating data in the present study were very simple. They can be described in order of the topics of selection, populstion, instrumentatioh, and evaluation of responses, which immediately follow. Selection: A letter was sent to each of the fifty-five fifth grade teachers in the hetropolitan Schools of Muncie, Indiana. p~rpose of the study and asked for their cooperation. (See ~nclosed card. This letter contained an explanation of the s;;c~~iv) in each letter was a stamped, self-addressed reply On the reply card the receiver was to indicate his willingness to p~rticipate. If the teacher ~ished to participate, he consented to a ten-minute interview. The teacher selected three of his students for individual interviews of approximately equal length. The letter requested that the teacher select a student who, in his estiffiation, was a "good student", another whom he felt liked social studies, Bnd a third student who was a "poor st1.Adent" in his estiffiation. The students were chosen in this manner in order to obtain a cross-section of the students in each individual classroom. If a teacher returned a positive 16 reply he was so that c0nt~cted 2 time ~nd for the ~&te interviews could to set. [cful&tion: t~e Fy of the interviewing period which 8nd extended frem J&l1.usry 23, lS6c through F'ebr:'1c.ry ,26,1<;62, fcrty teachers interviewed. :JDe hunr1red forty students hed baen :~nd Due to a d~screrancy in the format of the children's interview in the first three sets, where a set consisted of one teacher und three of his students, these sets were discounted in the evaluation of the The st~dy. sacple used in this study was thus reduced to 37 teachers and III students for Iurrcses of evslu9tion. re~resented the Muncie This s6~ple t\o.'enty of the twenty-two elerr:entary schools in Schools. ~etrop:lit~n Interviews conducted with each teacher ~ere student in the school Such buildin~. ~reas as Q ~nd each teachers' leunge, the nurse's cliniC, or the music room were used fur interviewing. nny room where it was ,elt that tje child would receive a utilized. ~jni~um of distr&cting influence was The length of an jntervi~w veried from seven to twelve rJinutes. Of the III students, 60 were boys Of the "]7 Hqood students The of the nu~ber ', r~cr st~dents of girls excee~ed c::: tegory Df the "g ood s t uden t InstrumentEtion: The 51 were girls. 12 "'"ere boys :.nc 25 1J.Jcre s~bject grou~, of the f:vorite fs~~le; l e~d 24 were m~le s~d l~ ~irls; were 24 were boys sod 13 were girls. the nurrber of boys only in tte I' . instr~~ent used with teoohsrs was a 17 one-pege interview guide. with the children. n two-pdge interview outline was used The items were a compilbtion of those suggested for use by various authors, since no single available instrument seemed to be valid for use with children in the measurement of their perception of classroom practices and atmosphere. following topics: Items on the teacher-interview included the (1) method used in teaching socia 1 studies; (2) resp:mse of students to preser,tation of subject metter; (3) neture and number of social studies projects for student participation during the school year; (4) emrloyment of bnd success with the problem-solving a[proach; (5) wi 11ingness to a t tempt "unusua 1 tI a cti vi ties ,~ bout which they n.ight feel "insecure" in terms of outcome; (6) excursions or field trips taken by the class during the year; (7) role playing and dralliatization used in social studies; (8) use of study guides by the tea cher; ond (9) rea c tion to a chi Id who is "individualistic and non-conforming". Items on the student-interview were related to the following topics: (1) freedom to disagree; (2) freedom to handle materials in the classroom; (3) student-centered or teecher-centered classroom; (4) strength of impression of ,,;hat is Leing studied; (5) response to the presentbtion of subject matter; (6) group work; (7) view of society; (5) divergent thinking; and (9) construction in the classroom. Evaluation of Responses: In the evaluation of each questionnbire tte raw score assigned to each teacher was determined by an equivalent to the number of positive responses he made on nine of eleven items included in the 18 teacher's ~ interviE~. mean of 4.5. A fossible range Two items included in the no variance in teacher response. student ~as ~as from 0 to 9 intervie~ ~ith produced The score assigned to each also determined by an equivalent to the number of positive responses on nine 3eparate items. A }earson ~roduct-~oment (r) ~ith .05 level of con- fidence w~s then used to determine three separate correlations: (1) the ra~ scores of the IIgood studentll with the of the teacher; (2) the student 11 ~i th ra~ ra~ score score of the IIfavorite subject the raw score of the teacher; (3) the raw score of tte "poor studentll wi til the raw score of the teacher. Individual items were examined for significant response patterns. Finally, a composite score ~as obtained for each teacher by adding the teacher's raw score to one-third of the scores for each set of three students interviewed per teacher. CR.,iTER IV The ~0~;ut3tion f~r of the Fe:rscn lr) e~ch ~f the three? e-rours of stucierts ':,1 th thei!' tes chen; !'eve:: ~ed the the 1'","Jd students II ,',ere c~T'rel[' t:::d 'di !' C' ',; S (' 0 !' e s :) b t ,<:,i rl e d fer t h 0 t esc be!' s; ra~ scores ott~incd TWG~ty-six c e rl t ~ methods in teaching or thirty the unit, in me~sure scci~~ ~er te3c~ing teacher ~ll .05 r no ton e, 'b 1) t st~~jes. • sccisl studies. i~rressi2n the level. 2. C 'J mt si~ty-twJ i rn t i:m 0 ~O~ 'f' The rem:injng eleven un 2n itc!::' CD>S ~ent, in 1f/h.ich rer cent, of the to llse~ t~~~rj scc13l 6:id detect s.n i~dlvjd~,jls "corr,rl.stely 3bsorb:::d il in their soci2l studies th,~ ~IJ h e !1 three coefficients of student 3ttitude strrcs;here of Ilsusrension" or hO\o.'ever, fel t .3 3 cent, rerorted the use :)f one method, studies fourteen, Gr thirty-eight :re'!" three, cr seeros r9'/J of the thirty-seven te::ch2!'s, cr seventy r or: 0 r t -2 d t he us e :) f te~~hers, (J) r fGr the tS2chers. found to he significsGt 2t the ~ere th the ',vc!'e worl~. thjrty-seve~ t8~chers, t their stUdents were !'r",rely or never 20 rr~je~ts, yielded twe r ty-thra9 r~sit1ve six t Y- bl D r e:' c e n t 0 f the t c : =1, ;:0 nrl f 0 1.1 r t e 8 n, eight rer cent, neg:tJve reSrcns2s. T'CS't-~/u4~ "nsc>' ........ _, t't,onty-three .,-,... ..... ,r, n(·'C1~ ''''"''". _.::,."" t; V"" - reSr2ns2s, tr~ j l' t Y- In the esse of the \~n"" ~.--. __ D:~ ~r ~b:ut 1I.1i....,p _.1._ (':"'_. 2h '-"u·t __ -....- v t,·'pntyw_ four ;.er cs!"t :)i' the tilirty-seven tesC''1ers rc;rcrt2c errrl::yi..'lg the rr~t18T selving ~rrr~Gch nIl inci:Lc3ted3 fC21ing ci' ,< in the they fe~t whi~h they c0nsj~ered 0" unusual activities in ~ut ~jtte~rted o?:' t!iirty-t\·;r:; :;'er r:-ent c:!" t":e Twanty-three re~ OT' ::-n 2ctivity Hbich they te:~.cheT's hc'.d cent o~ sixty-t~D the ~er t?~ch2rs hs~ fourteen or thirty-ei~ht Twelve rr?~1e2YCUrsi:ns twenty-fjve not dune so. cent oi the t:cchers the use of role-playin e or dr::u(,tization in re~~ining ~hich Tb? rerraining rem~injng the cbcut the Dutcome of 2ctivitips in their classroom. or sixty-eight cl~ssrccm. seventy-three rer cent of the tcochers rero!'ted th::-,t they hE,d n:-:t ~b8Ut att2~rtjnE insecure in terms of the outcome. twenty-seven studias "l'l'lcderate de€,ree of success." f0r cent of the teachers rerorted the c13ssroom scci~l SOCi:~ll rer~rted st')diesj reI' cent, hOwever, re,t:orte-::l th:Jt tilCY did not em:rloy either c;ctlv:lty in the nrcc of scci~l studies. Eleven or thjrty reI' cent cf the 21 teschers rerorted stu~y ~uides; th~t ttey ; the r c rr j On the f')11o\;dnz t3s~hers' cent cf the t8?chers ;21'" ::. n i n g t r: n::; r s; r 1" :) v: i rr co tel y t'.'! e n t y - s eve n j tern" hI8'1tY-:Gl)1" :Jr six-ty-fiv·:; per cent 0 teschers r2icrted that t~ey "f0el thrc3 toned 11 by suche:: chi 11. 8,... orde~ In c~ s",:"e;estions fr:::m the study cr thirty-five per cent of th I'::!islikc 'l extensive use sixt82n ::r forty-thrc3 1)sin~ g 11 ide s i I ~~je f'er ~ tesc~er scc;re on s11 iterr.s he ShOll1d: t:J have 3n entirely rositjve (2) use E c0mbinsticn methods in te:.chirr, scci::.,l studies; ~c) ce ~bl'? of to i!2tect En Co,tr::::sphere of Ilsus;ensicn on the r2rt :)f stlJde'1ts :)r ':Jf indivjdu~ls being currrletely sbsorbsd in tteir soc1?1 stllci'?s work "t tirres; (c) use ~r')jccts in his su::::i::l in the cl3ssro:)m ". . Uch he cunsiderad unlls'.1s1 :;,!1d :::b8 1Jt he m~y tri);:"s feel insecure in terms of C.S ~ le.s.rnir.r: ::2tivity; dram~tiz3tion ( f" ~ ) US,? field \2) em:;:lcy role rL:iying or in the s8cial studies cl?ss consirjst'ed Eppr:Jpris te; (h) study gvides; outc:)~e; ',\)~lich ~c(:d E r.d uc;e ~h2n they yere sugg~sti C'ns fr..:.m (i) hsve S ;;ositive attitude to\oJsrd the ,Iinc:!i vidu:; J j,stic S.!"ld n'")D-c:;nforming ii child. StuClent Intervi'?\I.': NinetV-.!"line stude!1ts or eightY-!1i!1.3 per cent of the students whc were interviewed gave f~ll 22 of ttejr ;r2soDt saci:l studies des~ri~t1c~s students e~2ve~ studying in W8r~ L i gh t 11 ~r 0 rer~ent soci~l f' t h'" hi C 1 v e \,'3 r stu~i~s enthusissm fer 30cisl not ~ble twelve to rec311 whst they : t the time of the e " roc I' S t Cn the 13020 student". ~ere wor~; 1) d e "1 t s ", t h:r e e j~terview. 1'.' e:r e item c::ncert'1irg the student I s se~~:)!'"l(:1 sixty-seven or sixty stu~ies, per~ent of the; stlJdents SE:td th:'t th::y (lid '!C't "excited" ?"h::::ut "Jhst th2Y studjed in sc;cisl sturlies, the r'2sairdnC' forty ;.ercent or forty-four' students ssid they h:::.d ilnever" been exci ted about soci~l studies durjng this schoel y~ar. Itew nU!'!lbers three ena f::::ur on the jnterview were :{llesti::-ns used divereently. ene ~C'rce~t, Cn the third item, fifty-fivs m~de res:~nses rerce~~ of the ver~ert rather then c()n~ernin§': land ~r ~1ver~e~t w~ich t~irteen were indjcstive :f thinkinq. E On thn f!"icndshir in or indicate~ thj~ty-t~:: n item f::rei?n see~ed m~re reSrOr'lSC's nG~r}y c~nverg~nt t~tnkjna. Thirty-six rercent of the students resrnnded cfftrm?tjv21y during this schoal yesr. eight rercent f~urth [crcent of the students ninety-seven st 1Jdents, which fifty- resrans2s indicative of con- ~~de the est2blishrr10 :1t cf f8u~teen resI0nses stu~e~ts ~h1ch stu~ents, m~de The rem3injng seventy-five or neE~tive resronses. ~inety-seven or si~ty- 23 eighty-seven ;crccnt felt free t: r~r0rtedly clessmstes ever tcrics under discussi~n 'I'h-== Y"2mainJng fuurt'?en stllcsnts tete;l rilj feel n:~t fifty-one rercent c frc2d:rr1. s~ch rerorte~ in social stu~ies. h:elve }:'2T'Cent CY' with dis~grec ~f the Fifty-seven students or thst they felt free to cxrlore end exreriment with things in the c12ssrcom; however, ftfty- students or forty-nine percent, did not feel this f~ur freed(~m. F'orty-fcur students or they did work j n s;rall ~roup g1"~ur: Sixty-seven ~ork. ,9.1 out forty I='cr(~ent s tn sod s 1 st'Jci as ~tudents c; nc s8id enj:;yed or sixty rercent, ssi~ that Cn the finsl item in t ho student they dicj enjoy grcup work. u~ intervj_2'.t!, seventy-six students or percent viewed their classroom 8S ,TrI~rC'ximstely sivty-nine being student-centered. The remaining thirty-fjve students, thjrty-one rercent jndicated th2t they In thou~ht their clsssr~om wss t83cher-centered. to the jnfarmeti0n ~dditi~n ~ained en each of the nine items, the data rev821ed that students Ierceived some indlvidusls 5S "helrful" Dnd lIencourc~ing'l. ~.tcut thirty-one rercent of the studects attrjbuted tho role of the most helpful rers8n to less thon !1!10 one d ~ne 3S E frienrl; sixty-eight rercent to the teacher; rorcent to "myself"; ::n:d one [crcent cJnsinered most helrfu1. In th~'? lstter cDse, thEt cf the mcst encour:ging flg:ure, forty rercent of tr:e students attributed this rc18 to the tee,cher; twenty-thrcE~ percent to s friend; 2nd thirty-seven r er'.;ent ccns i nered "myself" or " no one'l 2S most enccursgin€,. 24 Gomtosite Scores: Seventeen of the thirty-seven teachers or forty-six percent received composite scores above the mean of 9.8 while twenty of the thirty-seven teachers or fifty-four percent, received a composite score telow the mean. the distribution of the scores indicated one very creative teacher, nine moderately creative teaCllers, twelve less cre&tive teachers, and fifteen who used very little creativity in the teaching of social studies as evidenced by their composite score. No teacher was found to be cOffipletely lacking in crebtivity since the lowest composite score was 5.67. Summary 2nd Conclusions lhe of this study has been to investigate the pur~ose question - Do eleITientary teachers attempt to teach creatively? If so, to what extent? The area in which the investigation was carried out was fifth-grade social studies. Indivicuol interviews were conducted with thirty- seven fifth grade teachers of the ~uncie hetropolitan Schools and with three students from each of their classrooms. reachers were asked to choose one student whom they perceived E.S a I'good student l ' , one student who, it was felt I1liked socisl stu.dies ll , and one whom they feltwc;s a t1poor student. I' .-.t the end of the interviewing period an eva lua tion wa s n~C. de of the res:t:onses to the q ues tionna ires from thirty-seven teachers end one hundred eleven students. C&lcul~tion of the Pearson (r) revsbled correlations between the teacher scores and each group of students which were significant at the .05 level. On the item an~lysis of the teacher interviews, four of nine practices regarded as indicstive of creativity in teaching were not employed by a majority of the teachers. Twenty-three of the thirty- seven teachers felt that they could not detect an atmosphere of "suspension l ' indicative of who were completely absorbed in their social st~dies work. Twenty-eight of the thirty- seven teachers admittedly did not use the problem solving ar:proach in the social studies class. thirty-seven teachers h~d Twenty-seven of the not attempted anything ~hich they considered unusual or in which they were unsure of the outcome. fwenty-five of the thirty-seven teachers, for whatever reason, have not made an excursion with their class this year. The following conclusions seem to be justifiable from the do. ta obtained in this study: (a) the classroom a tmo- sphere in social studies in a large numler of cases could be termed "inadeq ua te "; (t) there is an infreq uen t use of problem solving in social studies, the use of which is advocated for other subject areas; (c) practices of teac~ers lean toward the conventional and teachers seem reluctant to attempt anything which they might consider unusual. Teachers seemingly desire to feel secure in their teaching and are reluctant to try methods about which they might feel insecure or unsure; (d) a majority of the teachers in this study felt that the effort of making arrangements for a field trip are greater than the advantages of an excursion. It was concluded from the item analysis of the stUdent interviews that according to the students interviewed: Co.) Students do not receive a great deal of encouragement to apply divergent thinking to social stUdies problems; (b) Des pi te the ta ct tba t s tuderlts enj oy g roup-work in socia I studies, group work is not employed by teachers; (c) 27 Industrial arts, which involves construction activities, is a minimal portion of the social studies program. Do teachers element~ry att6m~t to teach creatively in the classroom, as indicated by their teaching in social studies( It would seem '~:ise to conclude from the scores the t the problem can not be answered "yes" or "no". No teacher was found to be completely lacking in the atility to teach creatively. ~ll of the teachers with the exception of one had at least two areas, however, in which definite could be made in order to improve thejr standing on ch~nge a creBtivity scale. lhe act of teaching creatively, then, c& n be put on a continuum from "highly crea ti ve" to extremely low in creativity, or "uncreative", with most teaching falling somewhere in between. The greatest number of teachers were found slightly below the mean. Suggestions for Further Research ~ number of interesting questions and points for future exploration were encountered durins the collection of data for the study. 1. ~hat Some of these were as follows: is the influence of race and socia-economic level on individual creativity? Is one race or socio-economic level more likely to engage in creative activities than another race or socioeconomi cleve 1 c( 2. What can be done to emphasize the creative process instead of only the end product so that stUdents 28 will recei\e a more accurate and direct picture of the tot&l 3. ~rocess? What are the challenges in creativity? How much emphasis should be placed on determining the instructional media that would be most effective in putting across the process? 4. Will students tend to be more flexible, imaginative, and creative if they are exposed to a wider variety of instructional materials? 5. Is an open system necessarily conducive to creative thinkicg? ~hat effect does an open system have on divergent thinking? 6. How can professional peo~le adapt the processes to increase divergent thinkicg to the elementary cIa s sroom~i 7. How successful are creativity workshops in improving a te&cher's ability to teach creatively as measured by increase in number of creative practices before and after the workshop? 8. To what extent do the findings of stUdies with essentially the sallie format, but different items on interviews or questionnaires, differ from one another'? For example, if the questionnaires in the present study consisted only of the very direct twenty ideas compiled by Torrance, would the conclusions be the same as were drawn in the present study? 29 With so many unanswered questions and so much promise of future development and need, the field of creativity has become a fascinating one for persons in a wide variety of professions. 30 FOOTiWTES lCharles S~mner Crow, Cre2tive Education (New York: Inc., 1937), p. 226. ~rentice-Hall, 2;;.lice Miel, Creativity in Teaching (Belmont, California: Wadsworth Publishing Co., 1966), p. 8. 3Jacob W. Getzels and Fhilip W. Jackson, Cre8tivit~ and Intelligence - 1xplorations with Gifted Students (New York: John Wiley & Sons, Inc., 1962), p. 14. 4hie 1, p. 164. 5~braham Shumsky, Creative Teaching in the Elementary School (New York: Af~leton - Century - Crofts, 1965), p. 53. 6Calvin W. Taylor and Faul Torrance, "Education and Creativity", CreativiiY: Frogre~~ and Potential (New York: McGraw Hill Book Co., 1964), p. 50. 7.r..dgar Dale, "Education for Creativity", The News Letter, Vol. XXX No.3 (December, 1964), p. 1. 8 Ibid • 9Reginald 1. Archambault, "Education and Creativity", Journal of General Education, Vol. XVII No.3 (October, 1966), p. 55. lCRobert C. Wilson, "Creativity,lr iducation for the Gifted, NSSE 57th Yearbook - Part 11 (Chjcago: University of Chicago Press, 1958), p. 108. llfi. Hurwi tz, ftCrea ti vi ty :~ues tions ,5nd .hnswers," School Arts, LXV (January, 1966), p. 16. 12£. Faul Torrance, "Creativity: What Research Says to the Tea cher. II 1~i:Ji., Dept. of CIa ssroom Tea chers American 1ducational hesearch hssociation, No. 28 (April, 1963), p. 2. 131)a Ie, p. 1. 14Archambault, p. 202. 15Torrance, p. 4. 16F'rederick S. Breed, Education and the New Realism (New York: The IvlacmillBn Co., 1939), p. 9l. In'del, p. 61. 31 18Murray J. Lee and Doris M. Lee, "Creative Experiences: l-'roviding for Creative Living", The Child and his Curriculum (New York: Appleton - CeQtury - Crofts, Inc., 1960), p. 503. 19Ibid., p. 501. 20 Ibid • 21 Crow , p. 227. 22ShumSky, p. 208. 23lbid. 24wi llia m C .....olf Jr., "A Survey of be lieLs and fractices Relative to Creativity," The Journ::.l of Educational Research, Vol LVII No.7 (t-iarch, 1964), p. 3d4. 25Edward .Fred Stone, ",h Guide to Frinciples of Creativity - In-- Teaching with Suggestions for Use in .l:;lementa ry Soch,l Studies (Fa rts I a nd II), II Disserta tion Abstracts, Vol. AXIl l-'t. 1 (July - October, 1961), p. 507. 26J esse ~. Bond, II l • na ly si s of Observed Trc::i ts of Teachers rtated Superior in Demonstrating Creativeness in Teaching," Journ&l of ~duco.l.ional Research, Vol 1111 i~o. I (September, 1959), p. 11. 27Ib"d ' __1_., p.o. 28Carol F. harshall, "Classroom Ferceptions of Highly Creative and Less Creative Teachers in the Elementary School," Dissertation abstracts, Vol. XXIII Pt. 4 (April - June, 1963), p. 369129Janet H. Gilbert, tlCreativity, Critical Thinking, arrl Ferforrr..ance in Social Studies," Dissertation Abstracts, Vol. XXII l-'t. 2 (November, 1961 - }ebruary, 1962), p. 1872. 3 0 Frank B. ~Iay, "Crea ti ve Thinking: A Fa ctoria 1 Study of Seventy-Grade Children," Dissertation Abstracts, Vol. XXII Ft. 2 (November, 1961 - February, 1962), p. 1&76. 3 1Donald J. Hadley, "Experimental Rela tionships between Creativity and Anxiety," Dissertation .f-bstracts, Vol. XXVI Pt. 3 (November - Lecember, 1965), p. 2566. 3 2David E. Day, "Crea ting a Clima te For Crea ti vi ty, It School and Society, Vol. XCIV (October 15, 1966), p. 330. 33Ibid. 3 4 Taylor, p. 98. 32 35Shumsky, p. 49. 3 6 Ibid., p. 55. 371'Uel, p. IdO. 38Lbura L.irbes, Spurs to Creative 'leaching (New York: p: 26. G. F. futne~'s Sons, 1959), 391' rb nk ba rron, "l< reeing Ca pa ci ty to be Crea ti ve, 11 New Insights end the Curriculum (v.;a shington: Associa tion ~upervision and Curriculum Development, 1963), p. 292. 4eIbia., p. 31b. 41Teylor, p. 99. 42Torrance, p. 17. 4-3 u"h urns k y, p. 1.7 '"t • 44Calvin w. Taylor, fossible fositive and Negative Effects of Instructional Nedia on Creativity (Salt Lake City: university of Utah, 1960), p. 1. 45iorrance, p. 16. 46Dale, p. 3. lil-'J-ENDIX 34 eva lua tion of the crea ti venES:-; (a ccoroing to a pa r ti Cllla r lis t of factors) of the three stud8uts interJiE~gd and his/her own descrip- tion of his/her own p&rticu]br aprroach to social studies. teachers will be asked the same questions. All of the The interview should last from five to ten minut2s. Us.§_oLlnfortl}§;.li.QrL~ (1) The purpose of this study is not to make judgments as to good or bad teaching in social studies. Its purpose is to determine the prsvclence of certain factors in social studies teaching that receive the label Ilcreative". (2) The names of teachers and students will not appear anywhere in the report. results of the study arc ~o The be ta"bula ted in a chi square table which of course presumes anonymity. Please notice at this time the enclosed card. Bfter marking it appropriately pleasQ drop it in a mailbox by Jonuary 20, 1968. Reply must be immediateJ Once again 7 your cooperation will be greatly appreciated. do not forget your sign&ture on the enclosed postcard. Sincerely, " Please described as whjch of the followinga" u.."lit b" do pro,ject soc:.1alized not anyone cLettlod} but e. problem solving c. (If ,you use the uni t af,prOB(.:h') combina tion of methods B Do YCYU feel. obliga ted to include every subject area in the t..m.i t? Do you ever d,etect an a tmospb6u.'(~ thel~ ";suspcnsion" or individua Is being 0<' completely ab~orbed examples,> Could you plea se desc:ei\"e HI€ soci£' 1 stl.ldiE!S projects in in the2u.C(~\:;~:;;:' How would you descd.bo Do you y~)U have hod in employing the problem a?proach~ e~er attempt 5ctivities in jour classroom which you consider unusual and ",bout which Can Explain--specific engaged'~ which your Glass has solving social studies work? 'y1:)1l tell ':u.e about an any role playing or i-:-l to:rm3 of ov,tcome you~ e:~r.ur;~:lon In what capacity have insecure? ycnlr class has made this year'? they have dl'am&.tiz~~ti('Hl fE~el en~aged About in? used s:tudy g',l1.des? yQU in; interviewed on & scale of 1-4 (one 1s Classify each of trle stuclent2 the highest rattng :lnd:!.e.atlng)oss'::tss:L:m. of this ch[;..racteristlc to a great degree) av b. c" do ver1)al facilil:y verbal flueuc,:, flexibili ty unique ~nd workable ideas when presented routine problsffis or1~inalitY--trodGce3 - tith What is your recjctlon to 6 interviewed to be described ue~ Q~ ch:LJ.( ~s 1.\fbo ebB this way? be ()escl'ibed tJS: quieseent'l \I/lth= 36 ne ·V·~$ ~.QM"'..r" ~,.!t1;rr} by }f'Ol~ .r.·r,:r~· .:J\:'m$('n~ t(; W(~:rk:~j1L' t TY !~orl~ '1:~";'; f)fT: I~gnir~? ..~{: "~~ • • t""''l~JV·", ;~:}-=-2f.r ;.r>:)~~2: t'c nC ,c, r~ ~. If Vf.~;;' 'co? 11e )'0 '\.1 f~;H"Jl fr*~1;" 'tv i';')q:la'r~'1 anc f ~,;q:"~,,~,,,,'i;n:,:'nt 111~1';~h th<.';' 1/';1,cHiSE";)OtU th:l.tS '"vn."rr,.1 ·'::S "'v 1"'iv(l, ... ",,,,,~<.,,," ".n_.....~. ~ ....~ .' y'1}\,~ .4r6 ttl) ~1·'!rar.i dSf:id~ ~'o,r .iftSt..!~t:f;.:'t·!on~r:: Y('.H1J'!H~:tf ro~: Io·~.! t·n C:'~{';(;J ii' €' ~,~_l', ;'i~l")'e "'::il~~ ),~·,u y,~~s, 'Fba? '~~~t1t~j _ _ _ .......lIo,oI.oo .._ _ ....... I· • thtng~~ ["'~; ~n(:,)ul"aged i>&<&- on your' o'!;rn in on -y",'.l;;;' "wn) X!J~H.lL\r('h 1~K" :t.r(~ ~"i}U allt,1w'ed hrH' ytH. \;jou1d 2:lko to d(l s.('Hlle~, thi rag? you enoouraged to ~re D~'\l YOUT cle.!u;rtt.at~gt a~k fH~k i.l! _ qUDSt~01S 10 ~ ~;.f ........._-..C?-',,'W .• 1n class? (1l:'.H~~;tjons )'08 j ves ---.~ Does anyone? __yes no n r~o , .... ...)' 6S Dc::qu. Y~H~r- lftaaohf:u" t;..",lj 'Dye cve~·:,··-<;i~.i~g she no -----.... ~HH!2 • _ _rlO 1.n the t(,ztbO~"5? --no Can Y"'iU t€tll f?~~lI.0·";;!..(jt);l>.~~ e-ach cthe<' hoft! thing 10 lrla~ls? .. -.)1 os ----.. niil . y~u think o:r :lael about s():~e"", \,ben you and yOUi' cla!tSl;1atl)S hr:tV·;j a problem to silve do you decidE what to do yourselj'e~ ~'r does your tl.'l~.{;h('il" decide fOJ' Y0lol? ---_. tea~her 37 BIBLIOGRAPHY Books breed, rrederick S. E,duca tioD a nd The New Rea lism. York: The }~cffiillan Co., 1935. Crow, Charles S. Creative Education. hall, Inc., 1937. New York: New Frentice- uetze Is, Ja cob W., G. nd J&ckson, 1 hi lip W. Crea ti vi tv and Intelligence - EXQlorations with Gifteq Students. New York: ~iley, 1962. Gowan, John C., ed. Creativity: New York: ~iley, 1967. Its ~ducational Implications. Gruber, Howard E.; lerrell, Glenni and Wertheimer, Michael. Contemlorary .fip:proaches to l,;reative Thinking. New York: Atherton Press, 1963. Lee, Nurray .I., and Lee, Doris t<:. The Child and His Curriculum. New York: hrpleton - Century - Crofts, Inc., 1960. Narksberry, h&.ry Lee. foundation Harper &nd how, Inc., 1963. Miel, QQ ~~lice. Creetivity in Ie&.(!r~ing. ~adsworth fublishing Co., 1966. Creativity. New York: Eelmont, California: Creativity in 'leaching: InvitGitions and Instances. belmont, California: Wadsworth Fublishing Co ~, 1961. farnes, 3idney J., and Harding, Harold F., eds. 2 Source Book for Creative thinking. New York: Chcrles Scribner's Sons, 1962. ho trick, Ca Lharine. wha t is Crea ti ve l'hinking. fhilosophical Library ~ssociaticn, 1955. 1\eed, New York: Develoring Creative Talent - ~'-I.n b.dventure in 'lhinkirrg. New York: Vantage Fress, 1962. r:;. Seidel, G. J. The Crj_sis of Creativity. Notre Dame Press, 1966. r~otre Dame, Ind.: ShumSh. y, nbraham. Creative fe&.ching in ihe Blementary School. New York: Appleton - Century - Crofts, 1965. 38 Srr.ith, JiSmeS .~. Creative Tes.ching of §Qcial ;3tudies in the Lleruentsry School. Doston: hllyn ond bacon, 196'/. Setti:lK Conditions for Creative Teaching in the 61ementE.ll School. boston: ./"l.11yn end Bacon, 1966. Tay lor, Co. 1 vin ',,>/., ed. CreE:. ti vi ty : Frogres s ~ew York: McGraw - Hill, 1964. & nd Foten tia 1. Torra nce, E. 1-8 ul. l;eve lafme:; t and .i..VE:. Ilia tion af B~corded Frogrbmmed Lxperiences in Creative Thinking. ~inneapolis: University of Minnesota Fress, 1964. Von Fange, iugene K. frofessional Creativity. ~nglewood Cliffs, New Jersey: Frentiee - Hill, Inc., 1559. iies ley, Bruce E. Teaching the Socia 1 Studies. D. C. Heath and Co., 1942-.----lirbes, Laura. Slurs to Creative Teaching. P. Futnam's Sons, 1959. bos ton: New York: u. Yearbooks barron, Frank. "Crea tive Vision and 8xpressi.on, I' New InSights and the Curriculum. 1963 Yearbook of the Association for Supervision and Curriculuffi Development. ~dited by ~"lexander rrazier. ~\ashingt()n, L. C.: b. S. C. D. Fublications, 1963. Franseth, Jane. "l'reeing Capacity to be Crec~tive,I' New Insights and the Curriculum. 1963 Yearbook of the Association for Surervision and Curriculum Development. bdited by alexander ~razier. W6shington, G. C.: ~. s. C. D. Publications, 196~. wilson, rio bert C. "CreCi. ti vi ty ,II Ed uca ti ::m f<ll: the Gifted. Fifty-seventh Yearbook of the Nationsl Society for the Study of Lducation, PErt III. ~dited by Nelson E. Henry. Chicago, Illinois:. University of Chicago rress, 195cL [-{eports }\nderson, Harold H., ed. Interdisciplinary ~~ymposia on Crea ti vi U. i~ew York: He rper fress, 1959. Conant, James B.; strang, huth; and Willy, Paul. Creativity of Qifted and Talented Children. Addresses to the ~rr.erican rissociation for Gifted Children, New York. New York: Dureau of Fublications, Teachers College, Columbia University, 19591 39 farber, Seyn:our l/~. c;n( wilson h. H. L., eds • .t-Ian. and CiviliZe tion: ~on[lict and Creb ti vi ty; .!l Sy muosi urn. l~ew York: Iv.clIrsw - Hill, 1963. Ghiselin, Brewster. The CrEative Frocess, ~Symposium. berkeley: University of California Press, 1952. laylor, Ce,lvin '.N. Possitle Positive and Negativ~ Effects of Instructionc.l .t-iedia Qll Creativity. h.SCD eND';.) Project. Sc,lt L&.r.:e City, Utah: NaticGal Educationsl l,ssociation }ublic~tion, 1960. Studies Bond , Jesse A. "t~na ly sis of Observed Tra i ts of Tea chers Rated Superior in 0err:onstrating Creativeness in TE?a ching. II J our!l~l of 1:;d uca tic'na 1 .Rese'a rch. Ll I 1 (3eptember, 1955), 7-12. "ri Ibert, Jc:net E. "Crea ti vi ty, Cr i tica 1 Thinki ng and Ferformbnce in tJocial Studies.'1 Dissertation ~bstracts. LIll No. 1 (~epteffiber, 1959), p. 11. Greif, Iv.:> F. "l-,n Inves tiga tion of the Effect of a Dis tinct L(Lpha sis on Crea ti ve ano Cri tica 1 Thinking in tea cher i.ducation." Dissertation Abstracts, XXII, rt. II (jovember,.1961 - february, lS62), 1076. Hadley, .0onald J. 'ILxperim~nta 1 Rela tionships between Crea ti vi ty and .i-.n:d.ety. 'I Dissert::. tion Abstracts, XXVI, (November, 1965 - December, 1965), 2586. H&rris, Richard H. liThe lJevelopment and Valj_dation of a Test of Creative Acility." Doctoral Dissertation Series, Fublica tion 1,0. 13, 927. Ann .~rbor, Michigan: University ~icrofilms, 1955. Harsha 11, C& rol F. "CIa ssroom F erceptions of Highly Crea ti ve end Less Creative Teachers in the £lement&ry School.'1 Dissert&tion ~bstracts. XXIII, Ft. IV (April, 1963 June, 1963), 3691. N&y, rrL.nk b. "Creative Thinking: l'. Bacterial Study of Seventh-Grade Children." 1issertbtion ~bstracts. XAII, ft. 2 C~ovemter, 1961 - February, 1962), 1276. Stone, Edward F. "n GiJide to frincirles of Crea ti vi ty - In teaching ~ith 3u;gestions for Use in Element&ry Social ,studies. I! Lissert&tion fibstr&cts. XXII, 507. liJodtke, K. E., and ""allen, N. E. "Teacher Cl8ssroom Control, Fupil CreatiVity, and fupil Classroom Dehavior." Journal of .t;xperimental Education. AXXIV (Fall, 1965), 59-65. 40 l\rticles ... rchamba ul t, Regina Id D. "Ed uca tion and Crea ti vi ty. " Journal of General .c;ducbtion, AVIII, No.3 (October, 1966). Earron, It'rbnk. I'Creati'vity: What Research Says Jibout It." N£J-"> Journal, L (Barch, 1961), 17-19. Dole, Eagar. "Education for Creativity." The News Letter, Vol. 30, No.3 (December, 1964), 1-3. Day, Do. Vid. B. "Crea ting a Clima te for Crea ti vi ty. I' and Societ2, ALIV (October 15, 1966), 330-4. School l<orslund, Janet.l:.:. "Inquiry into the Nature of Creative of Creative Teaching." JO!!rnal of Education, XIV (h~ril, 1961), 72-82. Hallman, R. J. "Commonness of Creativity." Educational Theory. Vol. 13 (rlpril, 1963), 132-6. harris, Kenneth. "Don't Stifle Your Students' Nonconformity." NLi> Journal, LV (October, 1966), 24-26. Hearn, N. E. "Change for the Sake of Change: Why Not'( fHCE; l-'rojects to Advance Creativity in Education. I' N&tional Elementbry Frincipal, ~LVI (September, 1966), 50-2. Hughes, L. E. lI~duca tion for Crea ti vi ty ," Overview, I I (July, 1961), 33-9. Hurwi tz, f>.. "Crea ti vi ty : "iues tions and Answers." Arts, LAV (January, 1966), 30-3. School Lessinger, L. "Characteris tics of Crea ti ve Tea chers. " Journal of Secondary Education, XAXVIII (Ns.rch, 1963), 182-6. "r os tering Creb ti vi ty ." LAXVII (l-larch, 1967), 97. :r.:a rksberry 1 H. L. ]Lns tructor, Peabody, Scholl S. "Cultivation of Creativity." of Educ&.tion, ALIV (J:.;arch, lS67), 282-5. rlhodes, 1vlel. I'.r'.n Analysis of Creativity." Vol. 42, No.7 (april, 1961). Journal Jhi Delta ~~, Rubin, 1. J. "l ea ching for Crea ti vi ty . II Ca lifornia J ourna 1 of Secondc~ EdUCation, AAXV (December, 1960), 4l6-91. 41 Se linger, A. W. "f-roj ec t tipproa ch to Identifying and Encouraging 3tucent CreativitY,1I Californi~ .education. (February, 1966), 11-12. ,st&nistreet, G. Iv:. IIReal Hearling of Creativity," llJ.. (September, 1960), d. Instructor. Taylor, C. W. III::ffects of Instructional Media on Creativity." Educational Leader, AlA (npril, 1962), 453-56. 'l'rabue, 1'1.. 1,.• "Observations on Creativity.1I }orum, X.iVII (November, 1962), 55-65. ,~eir Educational , E. C. "Choice and Creativity in Teaching and Learning." Phi Delta Kappan, ALIII (June, 1962), 408-10. Wilson, .n. C. "Developing Creativity in Children." Educator, 11..11..)(.1 (September, 1960), 19-23. """olf, 'vdlliam C. "Survey of Beliefs and Frsctices helative to Crea ti vi ty. II Journ& 1 of Educa tione> l Research, L'J.VI1 (March, 1964), 304-85. 'Welman, tI.. IICultural F&.ctors and Creativity." Journal of Secondary Education, AXXVII (December, 1962), 454-60. Tape Becordings Sornson, He len H. "i-.re You a Crea ti ve lea cher? II Commencement Semin&r, ball State Jniversity. fhonotape recorded in Muncie, Indiana, June 3, 1961. ball State University, Teaching Materials Service. Sornson, Helen H. "Interes t and Crea ti vi ty. 11 } honotape recorded in Kuncie, Indiana, 1'I.pril 14, 1964. Ball .state University, Teaching Eaterials SE~rvice.