40 / Farough Abed
Schramm, W. What the research says. In W. Schramm (ed.), Qutdity in instructioncd tele-
vision. Honolulu: The University of Hawaii, 1972.
Seidmann, S.A. On the contributions of music to media productions. ECTJ, 1981, 29, 49-61.
Vinovich, G.S. The communicative significance of musical affect in eliaing differential
perception, cognition, and emotion in sound-motion media messages. Unpublished
doctoral dissertation, University of Southern California, 1975.
Washco, A. Ihe Effects of Music Upon Pidse Rate, Blood Pressure, and Mental Inmgery.
Philadelphia: Temple University, 1933.
Yarbus, A. Eye Movements and Vision. N.Y.: Plemun Press, 1967.
Zimny, G.H. and Weidenfeller, E W. Effects of music on GSR and hcart rate. American
Journal of Psychology, 1963, 76, 311-314.
Direct reprint requests to:
Farough Abed
Instructional Systems Technology Rm 210
Mitchell Hall 126A
School of Education
Indiana University
Bloomingion, Indiana 47405
Int'I J. Instructional Media Vol. 17(1), 1990
THE EFFECT OF VARIED
INSTRUCTIONAL STRATEGIES
(VISUAL AND VERBAL) IN
COMPLEMENTING PRINTED TEXT
MING-DER WU
Associate Professor
National Taiwan University
FRANCIS M. DWYER
Professor of Education
The Pennsylvania State University
ABSTRACT
The purpose of this study was to measure the relative effectiveness two types of illustrations
in facilitating college student achievement of different learning objectives. It also attempted to
evaluate whecher illustrations were able to improve learning when students were reading
textbook-like material which consisted of keywords in English. One hundred and ninety-eight
students at National Taiwan University participated in the study. Students took four criterion
tesrs immediately after receiving the respective instructional units. The results of this study indi-
care that a) the use of illustrations to complement print instruction did not automatically improve
student achievemenc on all Lypes of learning objectives. and b) variation in the amount of realistic
detail in illustrations did not make a significaru difference in facilitating student achievement
on the individual criterion tests.
INTRODUCTION
Visual illustrations have been commonly used in many instructional settings,
from kindergarten to college, to complement print instruction. It seems that the in-
clusion of illustrations in textbooks is based on the assumption that illustrations are
effective in facilitating student achievement on learning. In general, illustrated text
requires relatively more time and is more expensive to produce than text-only mate-
rial. Therefore, it is important to investigate the effects that illustrations have on
aiding learning.
42 / Ming-der Wu, Francis Dwyer
Rcscarch on ihc cftccts of illustrations has resulted in t,lixccl and contradictory
findings. By reviewing 55 illustrations-rclatcd experimcnial studies, Lcvie &
Lcntzlll concludecl ihat illustraiions could improve lcarning of print text, but the
cflccts, ,1. illustrations depended on how thcy wcrc used. In an analysis c,1. 74 studies
on visualized and conventional instructic,n carricd out in college classes. Cohen.
Ebcling. and Kulik {2 } reported that visualizecl instruction had no significant effects
on studcnis achicvcment. In a report of the rcsults of a systematic cvaluation on vis-
ual learning. Dwycr[3] concluded ihat " . . . tor specific cducational objectives
visualiiation of content material is no more ctlcctivc than the x: , mc instruciion
without \Pisualizatioii'(p.l).
On the other hand. illustrations consist of many types whcn used in textbooks.
Dwycr [4 ! listed eight types of illustrations based on the amount of realistic detail
possessed by the individual visualii_cd ircatment, and he pointccl out that an in-
crcasc in realism in a visual did not always cause a significant incrcasc in learning.
lt is obvious that if the effects ot'illustrations arc to bc stLidicd, the rcatism charac-
tcristics of illusirations should be taken into considcration.
Since 1965. Dwyer ancl his associaics conducted Il,, ,rc than 100 syslcmatic rc-
search studies [4.5.61. In these studies. a 2()OO^vord instruccional unit describing
the human heari. its parts. and its intcrnal process was dcvclopcd. A fter receiving
different types of prcsentation. ilic stuclcnls rcccivcd diflcrcnt critcrion tcsts. In a
report ot' cwerall studies. Dwycr[71 concludccl that all typesoI' visuals wcrc not
cquaIIV cf tcctihc in facititating studcnt acliievcmcnt of diflcrcnt cdticational objccP
tivcs. The present study was dcsigned lo dctcrniine whcihcr ihc s:tnic rcsulis‘,叫uld
be tound xx-hen Dwyer's instructional unil was translatcd i m[《 , (.hincse. Ilc)wcvcr,
quite a fc\v texrbooks used by college students in Taiwan arc in F.nglish. anci it is
not uncommon that in many cases students need to nieinorizc kcywords in the tcxt
in English. Under these circunisianccs. studcnts must undcrstaml(hc I,’ canin:[19‘ ’
keywords firsiin order to have a compleic "ndcrstancling《 、f xI. 'rw smdy  ̄
aticmpted io dctermine whether illustratiuns can aid lcarning c,r tcxt whc" kcy-
u,ords in the Iext are in English.
Related Literature
The et'tectiveness of illustrations has reccivcd wide atteniion in rcccnt clccadcs.
The advantages of illusirations in i1、、 prtwing lcarning wcrc found in many stuclics.
Peeck [8] indicaied thai illustrations thcilitate the learniiig of s ̄e spccifiablc in-
formation in a text. Florence & Geisclrnan[91 suggested t1、 “ t illustrations with
iconic symbols were easicr to locate and easier to recall from memory. Anglin l101
touncl that. in both immediate test ancl clclaycd tcsting. studcnts whl, rcad illusinitccl
prose materials achicvcd significantly highcr scores ihac tlic)sc siujc"ts xvho rcad
ihc samc prohe materials withoui ihc iIIListrations. IJcvinllll li,, “1 cighi assumcd
functions of illustraiions: decoration. rcn cration. mc)iivaiion. rciicration. rcp-
rcscnlation. organizaiion. intcrprctarion. and iranstornlatic)n . I.cvic & I.cnti [ lI in-
dicatcd the possiblc functions ot' inustraiionh as:‘l{Icntional. allcctivc.cl、9”itivc.
Varied Instructional Strategies , 43
and compensatory. Other research has shown that visuals also clarify and reinforce
printed instruaion and assist in understanding relationships and abstract concepts
necessary to specify learning objectives.
Methods and Procedures
Subjects of this experimental study consisted of one hundred and ninety-eight
undergraduate students enrolled at National Taiwan University. Twenty-six of the
subjects were solicited from the College of Science, while the rest were from the
College of Liberal Arts.
44 ' Ming-der Wu, Francis Dwyer
The content material employed in this stucly was a Chincsc t r;Lnslatic,n with some
modit'ication of Dwyer's insiructional unit. IL was a 2000-word insiructional unit
describing the human heart, its parts, and the proce.sses whicl] occur during the sys-
tolic and diastolic phases [4,5l. The content material was selcctcd spccifically bc-
cause it provided a hierarchy of several typcs of learning objcctivcs, extencling f rom
the learning of basic facts, concepts, etc. to complex problcm solving.
All students participating in this study received thcir rcspcccivc instructional
unit in textbook-like format. Students were randonlly assigncd to onc of five treat-
ment groups. Siudents in Treatment I, the control group received Chinese tcxL with-
out illustrations. Students in Treatment Group II received ihc sanic instruccional
text which was complemented by twenty simple line drawings of'thc heart which
were illustrating basic information presented in the text. Stuclents in Treatment
Group III received the identical texi; however, their instrtictional text was com-
plemented by twenty detailed shaded drawings. Students in Treatment Group IV
received a text only version of the instruction, however, approximately 200 relevant
English keywords were embedded in the text. For examplc, whcncvcr the worcls
"aorti' was mentioned in a sentence, it was spelled in English while the other words
in the sentence were in Chinese. Studenls in Treatnlent V received the text com-
plemenrcd by simple line illustrations along with the same English keywords rece-
vicd by studcnts in Treatmenc IV. Studcnts in each treatmc"t group were allowed
to intcract with their respective instructional unit for a long as they lelt necessary
to acquirc the intormation being prescntcd.
Criterion Measures
Immediately after receiving the instruction, students(c,c,k n、ur sclf-paccd critc-
rion tests. They were allowed to诅 ke as much time as nccdcd io coinplcic one
criterion test before proceeding to the ncxl. Studcnls in all fivc trcaf mcnt goups re-
ceived identical test items. Since the instructional units for Trcatnicnts IV and V
consisted of English keywords. the test items tor these two groups tiscd English
terms whenever appropriate. For example, when a keyword tcrms such as 'tricus-
pid' valve appeared in test items for Treainlent Group IV and V. it was spcllcd out
in English while other words in the sentence were in Chinese.
Tcsi ilcms were clcsigned to evaluate student achievcmcIl[1,1' spccific learning
objectivcs. Tcst I. Drawing Tcst. was to evaluatc student k”c,wlcdgc《、1. spcciric lc、ca-
tions of the parts of the heart. Students were required to draw a diagram of the heart
and place the numbers of the listed parts in their respective locations. Tesc 2. the
Identification Test. consisted of 20 multiple-choice items to evaluate studcnc abiliiy
to identify numbered parts on a Iine-drawing of a heart basecl on inll、rmation rc-
ceived from the instruction. Test 3. the Terminology Test. consistcd of 20 multiple-
choice items to measure student knoxvlcclgc of specific terms. dcl'initions, facts
rclatcd to the heart. Test 4. the Comprehension Test. also consisicd of 20 multiple-
choice items to evaluate student undc rstanding of the heart, its p:Erts. and its intc rnal
activities during the systolic and diastolic phases. Scores rccciveLI c、u ihcsc tour
tcsts xx;crc combined in an 80-itciii Total Criterion Tcst, which was clcsignccl io
Varied Instructional Strategies / 45
measure studenrs' total understanding of the concepts presented in the inscructional
material.
Each correctly answered item was scored as one point. The reliability coefTi-
cicnt tor each test employing Kuder-Richardson Formula 20 was: Tcst l, r=0.86;
Test 2, r=0.84; Test 3, r=0.81; Test 4, r=0.74; Total, r=0.94.
Results
Significant differences were found to exist among the means on four of the five
criterion tests (Drawing Test, F=10.51, df=4/193, p<.001; Identification Test,
F=9.08, df =4/193, p < .001; Terminology Test, F二7.04, df =4/193, p < .001; Total
Criterion Test, F二 7.86, df =4/193, p < .001). Insignificant differences were found
to exist in the comprehension test. The Scheffe Procedure was used to analyze dif-
ferences among the means. The alpha level of significance was set at the .05 level.
On the Drawing Test, students receiving the simple line treatment (Group II)
and detailed. shaded drawing treatment (Group III) achieved significantly higher
scores than students receiving the treatment without illustrations (Group I and
Group IV) (Table l).
Table 1
Scheffe Procedure Showing Mean Comparisons: Drawing Test
----
Treatment
Group N Mean S.D. Results
----
38 7.94 4.62 11>1
11 40 11.67 6.14 II>IV
111 41 11.97 4.82 111>1
IV 40 6.32 3.89 III>IV
V 39 9.33 3.66
----
On the Identification Test, students receiving the simple line treatment (Group
II) and detailed, shaded drawing treatment (Group III) achieved significantly
higher scores than students receiving the treatment consisting of English keywords,
but without illustrations (Group IV). Students of Group III achieved significantly
higher scores than students receiving the treatment wichout illustrations (Group I)
(Table 2).
46 / Ming-der Wu, Francis Dwyer
Table 2
Scheffe Procedure Showing Mean Comparisons: Identification Test
Treatment
Group N Mean S.D. Results
I
1。
V
12.23
15.22
15.73
10.82
4.31
4.46
3.16
4.83
II >IV
III > I
III > IV
V 39 13.28 4.57
--
--
On the Terminology Test, students receiving the simple linc treatment (Group
II and Group V) achieved significantly higher scores than students receiving the
treatment without illustrations (Group I and Group IV). Students receiving detailed
shaded drawing treatment (GrouplII) also achieved significantly higher scores than
students of Group I. (Table 3). On the Comprehension Test no significant differ-
ences were found between students receiving the varied treatment.s (Table 4).
Table 3
Scheffe Procedure Showing Mean Comparisons: Terminology Test
-----
Treatment
Group N Mean S.D. Results
--
--
38 10.55 3.88 11>1
11 40 14.25 4.70 II>IV
111 41 13.68 3.50 111>1
IV 40 10.82 5.14 V>l
V 39 13.76 3.43 V>ly
----
Table 4
Scheffe Procedure Showing Mean Comparisons: Comprehension Test
 ̄---
--
Treatment
Group N Mean S.D. Results
---
38 10.10 4.03 An
h Analysis of vari-
11 40 12.30 4.32 anceindicated
111 41 11.50 3.83 notwogroups
were significantly
IV 40 11.12 4.21
different at the
v 39 12.48 3.73 0.051evel.
----
Varied Instructional Strategies / 47
Table 5
Scheffe Procedure Showing Mean Comparisons: Total Criterion Test
Treatment
Group N Mean S.D. Results
40.84
53.45
52.90
39.10
48.87
14.28
18.02
13.14
16.54
12.59
II >l
II >IV
III > I
III > IV
On the Total Criterion Test, students receiving simple line treatment (Group II)
and detailed, shaded drawing treatment (Group III) achieved significantly higher
scores than students receiving the treatment without illustrations (Group I and
Group IV) (Table 5).
The main results were concluded as follows:
l. The use of illustrations to complement textbook-like instruction did facilitate
student achievement of most but not all of the learning objectives.
2. The effects of illustrations diminished when students received the treatment
which contained English keywords. With the exception of the Terminology
Test, no significant differences were found to exist among students receiving
the simple line treatment-plus-English Keywords and students receiving the
treatment without illustrations.
3. Variation in the amount of realistic detail in illustrations did not effect student
achievement.
Discussion and Conclusion
Illustrations were found to be effective in facilitating student achievement on the
Drawing Test. This finding supports the stimulus generalization theory [16] which
indicaces that when the test situation is similar to the learning situation, maximum
learning is achieved. On the other hand, the Drawing Test evaluated student knowl-
edge of spatial information provided in the illustrations. The exposure to illustra-
tions during the instruction becomes a substantial factor for learning the infor-
mation.
Illustrations were also found to be effective in facilitating student achievement
on the Identification Test and Terminology Test. Students required verbal and spa-
tial information to achieve on those two criterion tests. Illustrations and the key-
words accompanying the illustrations provided appropriate stimuli.
The Comprehension Test was designed to evaluate students' understanding of
complex procedures and processes that were less easily depicted by illustrations.
c
0
0
1
0
c
n
3
4
4
4
3
—
¨
…
Ⅳ
V
o
0
0
、
0
3
4
4
4
48 / Ming-der Wu, Francis Dwyer
The test emphasized mainly on students' verbal information obtained from the in-
structional unit. The advantages of illustrations disappearecl on che Comprehension
Te st.
However, the effects of illustrations on learning were not found when students
receiving their treatements consisted of English keyword. English is a second lan-
guage for Chinese students. It seems that those English keywords related to the
heart were difficult for them to comprehend. One possible explanation of the loss
of advantages of illustrations is that information conveyed by the text was complex,
thus requiring students to interact more with that information. Under those circum-
stances, students needed to sustain their attention in order to process the complex
information and paid less attention to information prc)vided by illustrations.
Simple line illustrations were found to be as effective as detailed, shaded draw-
ings in facilitating student achievement on the different criterion tests. This result
seems to indicate that simply increasing the realistic detail in illustrations does not
automatically increase its instructional effectiveness. The assunlption of realism
theories that teaming will be more effective as the number of stimuliin the learning
situation increases is not supported by this study. One possible explanation is that
the difference of he degree of realism between simple line illusiration and detailed,
shaded drawing is so small that no significant difference cc,Lild be R, ,1nd when ern-
ployed in instruction.
The results of this study indicate that visualizarion is an importanr inscructional
variable in improving student achievement. In this stucly visualization was used as
a rehearsal strategy to focus student attention on the critical learning attrihutes and
to allow time for the incoming information to remain in short-ternnnemory long
enough to be elaborated upon and encoclcd for long-terni menmry. Anclerson [23l,
Murray & Mosberg [24], Atkinson & Shiffrin [25] and Linclsay & Norman [26]
have comended that the longer an item of information is maintained in short-term
memory by rehearsal, the greaier the probability that it will be retained for re-
trieval. The different visual rehearsal strategies did not promote differential learn-
ing. Even though the visuals increased in realistic detail from simple line drawings
{o detailed shaded drawings may not have been sufficient to instigate increased
levels of information processing prerequisite for students to ef fcctively interact with
the instructional content and subsequently perform on criterion tesls measuring
achievement of the more complex educational objectives.
Specifically, the study supports the finding that illustrations are differentially
effective stimuli in facilitating student achievement of different objectives. How-
ever, the effects of illustrations need further study when the instructional unit con-
sisted of keywords which students have difficulty to comprehend. Different presen-
tation methods may be needed to aid the learning of cliffercnt educational
objecrives. The resuhs of this study indicate: first. the use of illustrations tc, comple-
ment self-paced print instruction does not automatically improve siudent achieve-
ment of all types of learning objectives; and second, viiriation in the amotint of
realistic detail in illustrations does not make significant diffcrenccs in facilitating
student achievement. Additionally, on inspection of rhc mcan achievement on the
Varied Instructional Strategies / 49
instructional criterion tests indicates that visualizations can significantly improve
student performance when used to complement print instruction. However, the use
of visualization alone does not maximize student achievement. Apparently, visual-
ized instruction needs to be complemented with additional kinds of learning strate-
gies which willenhance the rehearsal function and move greater quantities of infor-
niation from short-term into long-term memory. Additional research is needed to
examine the instructional effectiveness of varied types of learning strategies as they
impact singly and in combination on various types of visualized instruction.
REFERENCES
l. W. Levie and R. Lentz, Effects of Text Illustration: A Review of Research, Educ. Com-
mun. & Tech. 1., 30, 195-232 (1982).
2. P. A. Cohen, B. J. Ebeling and J. A. Kulik, A Meta-analysis of Outcome Studies of
Visual-Based Instruction, Educ. Co,nmun. & Tech. J.. 29, 26-36 (1981).
3. F. M. Dwyer, A Futurist Projection for the Program of S\Jstematic Ervduation. Paper
prcsemed at the 1982 AECI' Convention, Dallas, Texas (1982a).
4. F. M. Dwyer. Strategies for hnproving Visiial Learning: A Handbook for the Effective
Selection Design and Use of Visualized Materials. State College, PA: Learning Serv-
ices, Box 784, (1978).
5. F. M. Dwyer. Enhancing Visualized Instruction: Recommendations for the Practi-
tioner. State College, PA:, Learning Services, Box 784.
(i. C. A. Dwyer and F. M. Dwyer, The Effect of Visualized Instruction and Varied Rehear-
sal and Evaluation Strategies (Verbal and Visual) in Facilitating Students' Long-Term
Retencion on Tests Measuring Different Educational Objectives, J. Visual VerbaILan-
guazing, 5, 15-27, (1985).
7. F. M. Dwyer, The Program of Systematic Evaluation: A Brief Review. In7J Instruc-
tioiud Media, 10, 23-38 (1982b).
8. J. Peeck, Reteniion of Pictorial and Verbal Contenc of Text with Illustration. J. of Educ
Psvchol., 66, 880-888 (t974).
9 D. Florence and R. E. Geiselman. Human Performance Evaluation of Alternative
Graphic Display Symbologies. Percept & Motor Skills, 63, 399-406 (1986).
IO. G. A. Anglin, Prose-relevant Pictures and Older Learner's Recall of Writcen Prose
Echtc. Conunun. & Tech. 1., 34, 131-136 (1986).
Il. J. R. Levin, On Functions of Pictures in Prose. In F. J. Pirozzolo and M. C. Wittrock
(Eds.). Neurops)'chological and Cognitive Process in Reading, New York, Academic
Press. (1981).
1 2. H. T. DeMclo. Visual Sclt-Paccd Instruction and Visual Testing in Biological Sciences
at the Secondary Level. Unpublished doctoral disseriation. The Pennsylvania State
University. (1980).