DELBERT

C. MILLER
lateof IndianaUniversityBloomington

NEILJ. SALKlND
Universityof Kansas

50-

MEASUREMENT
6THEDITlON

~ Sage Publications

International Educational and Professional

Thousand Oaks London New Delhi

2~O1.-

Publisher

BASICGUIDEFORTHEDESIGNOF
A
SOCIALRESEARCH
PROPOSAL
----

11I

Certain elements must be included in the outline or plan of action for a social science research endeavor before the actual fieldwork or laboratory work is undertaken. Each of the individual guides in
this part of the Handbook of Research Des;gn and Social Measurement has been selected to assist in
planning the five main steps. Parts 3 (Applied and Evaluation Research), 5 (Guides to Methods and
Techniques of Collecting Data), and 6 (Guide to Statistical Analysis and Computer Resources) also
provide sets of guides that expand and further develop the main points listed here.
What follows is a point by point guide showing how to crea te an acceptable research proposal. You
can even use what follows as a checklist to help keep track of progress in your proposal development.
1. Identifying and Describing the Sociological Problem
A. Present a clear statel11ent of the problem, with important and relevant concepts defined
where necessary.
B. lIIustrate how the problem is amenable to being investigated through the use of a treatment
or test.
e.

Describe the significance of the problem with rcference to one or more of the following criteria:
1.
2.
3.
4.
.s.
6.

It represents an issue or issues that is/are particularly timcly in basic importance.

The question under consideration can be shown to relate to a practical prohlel11.
Outcomes may have significant implications for a widc audience.
Thc prohlel11at hand relates to an influential or critical population.
The completed work would fill a gap in the research literature.
Thc cOl11plctedwork would permit generalization to hroadcr principies of social interaction or general theory.
7. The dcfinition of an important concept or rclationship would be refined.
H. The results will have il11plicationsfor a wide rangc of practical problel11s.
9. Results in the past have been equivocal, and replication of the experimcnt is needed.
10. New or better insrrumentation for observing and analyzing data will be developed.
11. Opportuniries for gathering data, restricted hy rhe limited time available for gathering
particular data, are increased.
11. The Theoretical framework
A. Describe rhe relationship of rhe problel11or research questions to a theoretical framework.
1\. Del110nstrate how the problel11statel11entis rel\fed tu previous researeh ami how the previous rese:1rch narrows rhe current Sl'rof questions being asked.
111. The Hypotheses
A. Cll'arly stare rhe null ami research or alternare hypotheses. Presenr 1lrernare research hypotheses considerl'd feasihle within rhe framework of the theory.
1\. Dcfinl' rhe degree of risk you are willing ro rake by assigning a significance or 'Iype lleve!.
C. Define concepts or variables in operational terl11s.
l. Independent ami dependenr variables should bl' operationally dcfined.
2. The specific instruments used tu assess these variables should be identified and, if possihll', sal11plesincluded in the appendix.
.l. Psychol11etricd\faincluding reliability and validity 1boutthe instrul11entsshould be provided 1I1ddiscussed.
D. Describe possible shorrcol11ings ami their consequences.

. _.u

_.

14 ~ HAN1>BOOK Of RESEARCH DESICN ANI> SOCIAL MEASUREMENT

IV. Design of the Experiment
A. Describe the specific experimelllal desi~n, with partiwbr attentioll lo all variahll's allJ
whether the variables are tu be tested, controlleJ, or ignored.
B. Define threats tu internal and external validity anJ how they will be accollntcd foro
e. Spedfy statistical tests.
l. Specify the levc:lof confidence ;11which e;u:h IUIIIhypothesis will be testeJ.
2. Determine the power of tests, Kiventhe research hypothesis.
V. Sampling Procedllres
A. Describe the experimental anJ control samples.
B. Define and specify the population for which the research hypotheses are relevant.
l. Explain how sample size will be determined ami the type of sample that will be sclected.
e. Specify the nu:thod of drawing or sc:lecting sample(s).
D. Estimate rc:lative costs of collectin~ Jata Kiven the variolls sizes ami types of samples.
VI. Data Collection
A.

Estimate data collection time. If appropriate, be sure to specify

l. Number of questions to be asked of each respondent.
2. Approximatc time necdcd for intervicw.
J. Schedule as it has been constructed to this timc.

4. Preliminary testing of interview and results.

5. Alternativc plan for specifying adequatc time and place should 1-4 abovc go awry.
B. Inelude the following in description of intcrview procedurcs if appropriate.
l. Means of obtaining information (i.c., by direct intcrvicw, all or pan by mail, tc:lephonc,
or other means).
2. Description of special training for interviewcr.
VII. Analysis of Results
A. Specify method of analysis for each research hypothesis.
t. Statistical software packages to be used.
2. Use of tables, etc., with iIIustration by example.
3. Use of graphic techniques and tables that are to be used.
VIII. Discussion and Interpretation of Results
A. Discuss how data relate to hypotheses.
B. Discuss alternative plans for analysis.
C. Shortcomings of current research and possible solutions.
D. Ideas for future research.
IX.

Format and Publication or Reponing Plans

A. Proposal should conform to the style of the journal or institution for which it is intended;
usually, this will be the Publication Manual of the American Psycho/ogical Association (5th
ed., 200 t) or the American Socio/ogical Association Style Cuide (2nd ed., t 997).
B. Completion of final document according to department and graduate school requirements.
C. Selection of appropriate journals for submission with regard tu specific journal submission
guidelines.

..

FORMULATING
A
RESEARCH
PROBLEM
-_.-----...-------

The first step in the design of research is the selection of a question that has its roots in an interesting
thcoretical argumento The range of potential topics for social research is as broad as the range of social behavior, and although this fact does not make it easier for the researcher to make a choice, it does
represent a very large universe of ideas to explore. In addition, no researcher ever ignores his or her
personallife and professional experiences as the source for an idea that leads to a research question.

PART 2: BASIC RESEARCH DESIGN

The greatest of discoveries often begin as ideas formulated not in the confines of the classroom or lab.
oratory but on the trail during a walk or during the solitary early morning hours.
Selection of a problem represents a commitment of time, money, and energy, and it is not unusual
for a researcher to dedicate 6 months to ayear to finding a specific problem and formulating the question that will then lead to a research study. It may take many years to conduct the research and even
longer to see it through to publication.
The significance of a problem and the precision with which the research question is formulated
cannot be underestimated in the role they both play for helping to contribute to the body of knowledge in a particular field. The obvious question is "How can this significance be foreseen for research
not yet undertaken?" To a great extent, this is the challengc that faces all researchers and why the
quality of the research question initially asked is so important. The competent and experienced researcher is very familiar with the research literature and can identify what needs to be done next to
further answer the more general and theoretical questions. A rich array of theory and methodology is
available in which the proposed problem can be casto
The: student (or any re:seareher, for that matter) should be led through the process of identifying
interesting and potentially productive problems during training. The following set of general points
provide:s suggestions for how to find a research topie (including master's thesis or dissertation research activities) and the aecompanying research designo
1. Ide:ntifyinte:resting topics in the first se:tof courses that you take (which are usually core or required courses).
2. Ask que:stions and incorporate the answers ioto the body of knowledgc you acquire during your
courscwork.
3. Conside:r your own expcriences and personal eharaeteristies and qualities, sueh as your potential
professional growth and future carcer intere:sts, the ability and interests of the faculty with
whom you are working, your ability and desire:to work with a specific profcssor, and the avail.
ability of resources (money, equipment, ctc.) to complete your work.
As you continue: your coursework and begin to fee! more eomfortable as an aspiring rescarcher, be
sure to do the: following:
1. Enroll in se:minars through which you hope to expericnce intellectual growth.
2. Examine carefully how others (espe:ciallymore advanced graduate: students) have:approachcd
re:se:archproblems and the:steps the:y have:taken to get where they are in their careers.
3. Initiate small researeh proje:ctsin the:dire:etion of your intere:sts(usually under the supervision of
a faculty me:mber,cither as an original study or one relate:dto what the faculty membcr might be
doing).
4. Discuss sugge:sted topics with your adviser, examining whieh une:xplored areas of the ficld
should be:studie:d, what previous research has be:en done:, and what relevant literature bears
upon the specific question bcing asked.
5. Attend professional confere:nees whenever possible; although these might be in distant dties and
your resouree:s might be limite:d, re:member that graduate students for ye:arshave bcen sharing
rooms (and having a grcat deal of fun) and Ie:arningat the same time.
This set of suggestions assumes that the student takes his or her work seriously and is not simply
going through the motions tu "knock out a thesis,"IHlt instead is planning a scholarly career. From a
thesis may emerge publishcd anicles that will provide the base:for the researcher's reputation in the
ficld and the: springboard for future growth and contributions.
There are no shoneuts to becoming a eompetent rese,ueher. h involves a great deal of time and
practiee in e:vcrysense of the word. An increasing numbcr of expe:rienees in different scttings leads to
enhanced competenee. A suggested place tu stan, especially for the na'ive re:searcher, is a replication
of existing work. Sodal sdcncc needs this kind of resean:h badly in acquiring cumulative e:vidence:,
espedally when the results of previous rcsearch are equivoca!. The:srudent may utilize secondary data
tu cnrich rcscarch ami to minimin' expcnditurcs of time and other rcsourees.

15

16

HANDBOOK Of RESEARCH OESIGN AND SOCIAL MEASUREMENT

Several books can hclp novice researchers sclect important problems worthy of research, 'Iwo of
them are identifie:d bclow.

Further Readings on Formulating

a Rcscarch Problem

Cre:swell,John W. (1994). Researchdesign: Qualitative and quantitative approaches. Thousand Oaks,

CA: Sage.
Denzin, Norman K., & Uncoln, Yvonna S. (2000). Handbook o( qualitative research. Thousand
Oaks, CA: Sage.

ltII

EVALUATINGRESEARCH
STUDIES

An integral part of preparing any research proposal is reviewing the:existing literature. You'lIlearn
more about how to do this in Part 4 of the Ha"dbook o( Research Design and Social Measureme"t.
Here, the foeus is on the eriteria one should use when evaluating the research studies included in
one:'s review and used as a basis fur the:rationale for the:study une:wants to complete. To be an effective researeher, you ne:eda critical eye for the outstanding features as well as the deficits of any one
pieee of research.
The following list of que:stiuns shuuld hclp guide you thruugh the prucess of evaluating a research
report and in writing a literature review.
The Review of Previous Research
1. How cIusdy is the lite:rature reviewed in the study rclated tu previous literature?
2. Is the review recent? Are there any uutstanding rcferences (those of vital conceptual
significance) left out?
The Problem and Purpose
3.
4.
5.
6.
7.

Can you understand the statement of the problem?

Is the purpose uf the study dearly stated?
15the study's purpose tied tu the literature that is reviewed?
Is there a conceptual rationale to which the hypotheses are grounded?
Is therc a rationale for why the study is an imporrant one to do?

The Hypothesis
8. Are the research hypotheses cIearly and explicitly stated?
9. Do the hypotheses state a cIear association between variables?
10. Are the hypotheses grounded in theory or in a review and presentation of relevant
literature?
11. Are the hypotheses testable?

The Method
12.
13.
14.
15.

Are both the independent and dependent variables cIearly defined?

Are the definitions and descriptions of the variables complete?
15it cIear how the study was conducted?
Are there adequate reliability and validity data regarding the dependent variables?

PART 2: BASIC RESEARCH DESIGN

~ 17

The Sample
16. Was the sample se\ected in such a way that you think it is representative of the population?
17. Is it clear where the sample comes from and how it was selected?
18. How similar are the subjects in the study to those that have been used in other, similar
studies?
Results and Discussion
19.
20.
21.
22.

Does the aurhor relate the results to the review of literature?

Are the results related to the hypothesis?
Is the discussion of the results consistent with the results?
Does the discussion provide closure to the initial hypothesis that the author presents?

References

23.
24.
25.
26.
27.

Is the list of rcferences current?

Are the rcferences consistent in their format?
Are the rcferences complete?
Does the list of rcferences rcflect some of the most important rcference sources in the field?
Does each rcference cited in the body of the paper appear in the rcference list?

General Comments Ab(mt the Report

2M. Is it clearly written and understandaoll'?
29. Is the language unoiased (nonsexist :md relatively culture-free)?
.W. What arc the strengths and wcakncsses of the rcsearch?
31. What are the primary illlplications of the research?
32. Wh:1twould you do tu improve the research?

2.3.1 Useful and Beneficial Attributes

of a "Good" Research Problem
How does one judge the value of a rcsearch problem? Ask these seven questions as they apply to
the topics you have IInder consideration.
1.
2.

Isthe prohlel11concerned with hasic conccpts, such that the knowledge produccd will be cumulative :md ouild on an existing body of knowledgc?
Will the invl"stig:ttionof thl" reseuch proolcm rcsult in a testing of some theoretical formulation?

.~. Docs thl" fl'sl"arch prohll"111

allow for the cardul specification of the variables involved and use
of the l110stprecise and appropriate methods availablc?
4. Will an investigation of the research prohlem result in a contribution to methodology by the
discovery, devc:lopment. or rcfincmcnt of practicablc tools, techniques, or methods?
S. Will the research proolel11 utilize the relcvant concepts. theories. evidence, and techniques
from the discipline or suhdiscipline of interest?
6. Will thc integration of this single srudy into a planncd program of rclated research produce results that are lIIorl' IIIl'mingfulthan those achievcd prior to the intcgration?
7. Will therl' he opportllnity for the training of additional rcscarch scientists?

IN ~ HANL>800K Of RESEARCH DESIGN AND SOCIAL MEASlJREMENT

--

- --

ELEMENTS
OF RESEARCH
DESIGN

When selecting 3 prohlem for possihle research consideration. tlw complete research dl'sign and all
its dcmcnts must he considered and formally evaluated. Tahle l. J lists dimensions of the research design th3t nee:d tu he:considered throughollt the process of sclccting a prohlem. asking the qucstion.
framing the hypothesis. 3nd completing the experimento The following pages provide guides for
m3ny of the:se:design decisions. The choice of a rcsearch dcsign is of grc3t importancc hecausc it influcnces 311the outcomes of the srudy.
1.1111,.
'2.1

1\\I'OI~ JAN T F\C TORS IN DEHRMININC,

(H lilE IH:S[\RCH IH:SIGN

Rt's(>drch ftem('rJI

Ch();n',~

Type o( undcrlying Iheory

Gener.lllheory
Middle-range Iheory
Supposilions

Sludy design

. Pre-experimenlal (surveyl
Experimenlal
QUdsi-experimenlal

Access lo organizalions and

res~ondcnls

..
..

Requires

permission

o( inciividuals

Requires

permission

of

CH\R\CTERISTICS

orgdniLalion.lloffid.lls

Degree o( conlrol ovcr Ihe

social syslem heing sludied

. No conlrol
. Parlialconlrol
. Complele conlrol

Type o( dala availahle

. C.lse and ohservalional sludies only

Quanlitalive analysis only
Quanlilalive supplemenled wilh case ancl observational sludies
Olher (hislorical, cross-cultural, elc)

Temporal dimension

Cases (rom a single sociely al a single (Jerioo (cross-Sl'clionoll)

. Cases (rom a single sociely al many periocis (lime sNies or longiludinal)
Cases (rom many socielies al a single period (comp.ualive cross-culluroll!
Cases (ro~ many socielies al differenl periods (comparalive longiludinal)

Sample or universe lo he
sludied

.
..

Individuals in a role within a group

Pair o( inlerrelated

group

members

(dy.10)

Primary group (30 or less)

Secondary

group (31 or more)

Terliary group (crowd, public, etc)

. Slale, nalion, or sociely
. Single or few cases
. Small sample (under .lO)
large sample (more Ihan 30)

Sample size
" --

Dala source

Original

-_.Dala galhering melhod

--

.-

.
..
.
..
..

- -- - --

data lo be collected

by Ihe rescarcher

Archived

or secondary

da la in hand

Archived

or secondary

da la lo be collccleci

DireCI ohservalion
Interviews

Queslionnaire
Tesl or so me olher form of measuremcnt

PART 2: BASIC RESEARCH DESIGN

Research E/ement

Choices

Number of independenl
variables

.One
. More Ihan one

Number of dependent
variables

.One
. More Ihan one

Sclection of scales to Jssess

- dependenl_. variable
-_..Characteristics of dependent
variable(s)
Duration of sludy
Resuurces required fur
completion of sludy

_.-

Nominal
Ordinal
. Interval
. Ratio

Level of measurement

_._

.
.

Presence of reliability and validity data

-:'~sence_~f_reli~bili~a_~dvalid~ da~a___
Normally distrihuled
Nol normally distribuled

~ 19

-----_._----_.

Brief (less Ihan b months)

Long term (more than b monlhs)
Funds required
No fu~ds r~u~~d

SOURCE: Adapled in part from Matilda While Riley,Snci%gica/ Research: A Case Approach (New York:
Harcourt. Rrace & World. 19(3).

CHOOSING A RESEARCH
DESIGN

..

Empirical research in social science procceds in a variety of settings and contexts. Thc choice:of a design setting for any research project is generally of vital concern to the:re:searcher, who seeks to determine rhe validity of a hypothe:sis and how be:stto discover cvidencc to cithcr acce:ptor rcject it. Social
phenomena are almosr always complex, and control of the rclevant variables that contributc to that
complexity is difficult at best.
Tht' major question rhen becomes, "What design will best asccrtain associations or causal paths
among all the rclevant variables?" How the research design goes about answering th1tquestion may
well determine the future outcome of the study. It will most certainly determine the amounts of time,
money. and other resources required for the smdy.
The general guidclim' to '~srarrsrrong" should supersede any other design consideration. Evcry ef.
fort must be made ro sclect a design setting using a population characterized by a largeamount of vario
ability in both the independent and dependent variables. In general, the more variability, the more
likcly ir is rhar differences between groups will be detecrable. In addition, for any research pruject, inSl/ranCl' is importanr ami may be secured by combining case analysis with any other research designo
Intense examinarion of extreme cases at the tails of a distribution may reveal polari7.ed rclationships
rhar suggesr new hypotheses, designs, ami analyses of the data,
'I:,blt' 2.2 prt'st'nts a summary of the many different rypes of designs used in basic and applied rese:m:h. It describes rhe rype o( dt'sign, its cenrral characterisrics, \lid rhe ourcomes that are possible,
You may find many of rht'se similar, The one (or combination) you select will be a function primarily
o( Iht' nalme o( Iht' rt'st':m:h quesrion you are asking and rhe hyporl1t'sisthar reflecrs that qucstion.

20

HAND800K

I.lhlt, '!..2

OF RESEARCH DESIGN AND SOCIAL MEASUREMENT

lilE ()IHIIUN

Typeo( Research
Deslgn Settlng

------

1. D~crlpllvc surwy
a. Cross.sccllonal sludy
Examples: U.S. decennlal
census; lamt!s A. Davls,
UnderRraduale Career
Ot>clslons; Pelcr M. Blau &

O. D. Dunean, Thl' American

Occupallondl Strucrurl'
b. lonHlludlnal Sludy
Examples: Greg J. Duncan &
lames N. Morgan (Eds.I, Flve
Thousand Amerlc.1n Fdmilles

J ()ESICNS USE!) IN B\SI(" \N!) \PPUU)

Ct'ntr.ll ChiJf</ct('rislin

Educatlon, Occupatlon,
Earnln8s

dnd

J. Fleld sludl~
Examples: Robert lynd & Hclt!n
lynd, Mlddletown dnd
Mlddletown in Tr.1n.~ition;AURUSI
B. Holllngshead, Elmtown's
Youlh: The Impact o( Social
C/asses on Adolesu'n's dnd
Elmtown's Youth and C/m'own
Revi.llted; Wllllam F. Whyle.
Street Cornl'r Socie'y; f'hilllp E.
Hammond (Ed.), S(X'iu/o/lisb .1'
Work

()ulcolllt.'s

Cnnn'rned wilh infmm.ulnn gt.ner.lIly

nblalned by inlervlt!w m m.llllod
qUt'sllonnalre. Olher ~ourCt'S Ineludt!
offlcial reports or slalistles. Oceaslunally, dala banks nf uthcr rt!st!areht'r~
provlde appropriale infmn1.l1l0n.
Kt!qulrt.'5 an t!ffort tn procurt! 100%
enumerallon of Ihe sample under Sludy.
TIme serlt.'t are produeed shnwlng social
or behavloral eh.1ngl'5 over varyinH
perlods of lime.

A sl".hle

Dt!als wllh only .1 Ir.l("llon of .. Inl..1

n.I'., n1.lYIK' .m..lylt'd 1m slmpll'

rt'I.1Iinnships IK'lwl't.'n Iwn V.Uldhll'5.
Mulriv.Jri.Ut. .ln.1Iysis m.IY involve f.1clm
.lIldlysls, m.llrlx .m..lysl~, .md mulllple
disnimin.ml .,"..Iysl~.
"nlh qu.1nlil..llve .md qu.1Iir.11Ivt!d.lId .IIt'
.m.1IYled wilh ,.pprollri.ue IloIramelrk
m nonp..r.lmt!lrle mt!lhods.

-_._2. Sample survey

Examplt.s: Gallup, Harrls. and
Roper polls of publie opinion;
Cutrent Popu/atlon Surveys,
Bureau of Ihe Census; Wllllam H.
Sewell & Robert M. Hauscr,

1'f{)~pt'cliVt'

IHS[\f{("1t

pOpUI.11Ion.
Sampling melhods .Jre t!mployt.od In
pruvldc a samplt! Ih.11is an accuralt!
reprl'5t!ntallon nf Ihe lolal populalinn.
T~I hypolh~
may be t'Slablishcd. Tu
ensurc valldily, rese.ueher will ullli.lt!
It!ehniqu~ for sealln!\. p.IY cart!ful
anenlinn lo queslionnaire wmdlng .1nd
prt.'5enlallon, and inelude qUt.'5l1nns
on personal hackground and nlhcr
polenlially uselul v.Jriahles.
Conccrned primarlly wilh
proCt.'5st.'5and pdllcrns undl'r
invt!sliR.llion 01 a sinHlc !\fnup, f.llnily,
Inslllulion. CJr(anil.llion. m cOlllmunily.
Emphasls 15on Ihe sncial slruclun'
(Le., Inlerrelallnnships nf parts 01 Ihe
slruclure .md social inleraclion IdkinR
pl.lce).
Allcmpts nbservalions nf soci.1I
Inlcrdclions or invt!sliR.lles IhnrouRhly
Ihe rtociprocal pcrccplions and altiludt.'5
nf people playinR inlerdependenl
roles.
Dirtocl .md Ilolrlicip.1II1ohservalion,
inlerview, .1nd sc.1linl( 1t.'Chniques .\re
nfren employcd.

Usually refers lo relalively inlellslVl!

4. Case sludies of persons
Examples: Elizahelh Eddy,
,1nalysis of a sinl(le insldnce of a
phenomenon beinl! invt.'5lil(aled.
Becoming a Teacher: The Passagc
to Pro(essional S,atus; Erwin O.
Inv('Sli;alor inlervi('ws indlvidu.lls m
sludies life hislory documenls In ain
Smigel, The Wall Stree' Lawyer;
W. F. Conrell, The Railroadcr
insighl inlo behavior.
Allempls In discover unlqul' lelIureS .md
eommon Irails shared hy all persons
in a Riven classlfiealion.
Cas~ may he !\fouped hy Iype lo
diseover uniformilies.

vnlumt. nf inlmm.uinn

Ih.u Gm

ht! cI..sslflt!d hy Iypt., frt.'tlut!ncy, .md

mt!.uurt.'S 01 Cl'nlr.111t'ndt!nt.Y.
EKpense nf Ihe survey will he very large If
Iht' IKlpul.ulon Is suhsl.mll.d. Fln..1
ylt!ld: d.II.llh.11 m.IY he .m.lly.wd 1m nu.
merous rclallnnshlps.
Slandardl.led dala eapable of
cnmpar.11lvl! .malysls over suect.'5slve
lime Inlervals In whleh longlludlnal
sludles t'x.1mlnt! dge chdnges dnd
cros~-sl'Cllon.1I sludll'5 examine age
dlfft!rt!nCt'S.

1).11.1(.lIht.rt'd ('n..hll' m.my hYIKllh('",s

'0 IK' I('SIt'd Ih.u would nol h(.
.1I\\t'n..hll, lo lt'slillR wilh survt'y d.II.I.
Gr('.lIer cOlltrol is .I('hit'v('d hy focusinl(
on .1suhl(roup 01 I.lrger pOpUI.1Iion.
SocinlclRiC'al produ('ls such as
processcs, flolllt'rns, rolt!s, .IUilud(.s,
.md v.llu('s h(.!:tlm(' .IV.liI.1hlt,.

IJ.IIJ CJn he .lsSt'mhlt!d lo provide Insilu

in.o condllioninR 01 rt'lalionships .111d
causalive faclors.
I'erson.llily and S(K'i.lli,.lIion proCl'SSl'S
c.1n he idenlificd.
Cnncepls can he IcSlcd .md discnVt'red.
Cast.'5 c.m be CtKled .md slalislic..1 tesis
.lpplicd lo provide associalinns
helwcen v.1fiables.

PART 2: BASIC RESEARCH DESIGN

Type of Research
Design Settin/?

Central Characteristics

Prospective Outcomes

5. Comhined survey and case study ~urvey melhodolol\y is coml~~~~-wilh Exampll'.\: E. W. BurRess &
sludy of specific cases lo illuminale
leonard S. Cotlrell, Ir., PredielinM
relalionships firsl portrayl.'Ii in a
Success or Failure in Marriagc;
correlalional patlern and Ihen
Alfred C. Kinsey & Associales,
inlerpreled Ihrough casf' sludy to
Sexual Behavior in Ihe /-fuman
display processes and patlerns.
Male and Sexual 8ehavior in Ihf'
Cases are seleeted after a survey reveals
Human Female
Ihose Ihal Me hih or low on a
crilerion variahle or Ihose Ihal display
sinificanl characlerislics.
. Prediclion

studies

Cxampll's: Sheldon Glueck &

Eleanor Glueck, Pn'dictinM
Delinquency and Crime; Paul
Horsl, Thf' Prediclion of Personal
Adjuslmenl (see especially
Paul Wallln, "The Predietion of
Individual Behavior From Case
Studies")

RelalionShiPS accompanied
by process
and patlern data revealing personal
socialization in greater depth.
Two da la banks assembled: slatistical

Purpose is lo estimale, in advance of

participation, Ihe level 01 an
individual's performance in a iven
activity.
Search is made of a populalion lO find
faclors lo serve as basis for predlction
01 such oulcomes as success or failure

dala and case analysis dala.

Relalionships belween a number of

faclors and a prediction crilerion are
determlned.
Selecled faelors are weighted In the
construelion of pronoslic tables.
Pronoslic lahles are ulillzed lo make
prediclions.

in marriage, degree of success on

parole, findin polenlial delinquenls
al an early age, school success, or
criminal bchavior.
A discrele dependenl variahle is often
soughl, such as stable marriage vs.
broken (divorced) marriae, law ahld.
ing vs. criminal hehavior, delinquenl vs.
nondellnquenl boys and Iris, or
academic achlevers vs. academic
nonachlevers.

,--

~ 21

7. Conlrolle<!cxperlmcnls (wlth
major Iypes belng laboralory,
"natural: and field experlmenl)

a. laboralory
Cxample: Robert Bales,
f't'rsonaIiIY ilnd Inlf'fpt'rson.11

Behavior

Investigalor creall's .1silualion

wilh Ihe exacl condllions wanled
and In whlch he or she conlrols some
and manipulall.'S olher variables.
Investigalor nbservl.'S and melsures
Ihe elfecl of manipulalion 01
ind('pendenl variahles on depend('nl
variahles in a silualion in which olher
rel('vanl faclors are nmlrolk'li

h. "Nalural" expt'rlml!nh
(such a~ aoss,sc<:llonal

Relallonships found can be eonsldered

more precise as a result of eontrolling
"Inlerfcrin"variables.

ex pOSI faclo).
h.I/Jlpll',' 1'.S. Ch,lpln,
I xpt'riml'nlll fJt'siHII.\ in
Soclo/oHIc..1 Rt'~I..1reh

lo Fh.ld l'xJwriflll'nt
I x.I/Jlpll'; J. C. Milll'r.
hpt'riml'nh
in SIX'i.,1 I'ro( I'~,'

or, il

approprlate, Ignored.
Rewarcher capllallu'S on soml' nnlloln
chanes in Ihl' normal wmmunily
setlln and sludles Iheir e((('I" In m
experlnll'nlal de~lgn.
A Ir('alml'lIl nr social pmram may Iw
Iven lo 011I' rnup nf p('rMms md
Ih('lr personal adjuslm('nt compared
wilh that of ,1mup of pt'r1>ons wilh.
oul sUI'h iI pmgfilll1.
Ml1chlng 01 gmups m.,kt.s Ihe Iwo
roup~ homolll'nl'(lus whl'n Sl'11'('II'Ii
f.1I'lors 1ft.IlI'ld conSlanl hy mall:hln.
Involv('!> m.1nipul.llion 01 condlllons
by Ihl' I'xlx'rlmenll'r in mdl'r lo
dl'll'fmin(' possihll' CdUS.11f('I.llinn~.
Maxlmum volrio1llonIn Ihl' Indept.nd('II1
.1nd dl'f!t!ndl!nl volrlahlf'S Is hullllnlo
the struclure 01 thl' desln.
(xpt'rlmenl.ll and conlrol ruups are
t'slahllshed, holdln conslanl f,lI'lors
bellt'v('(!lo inlerfl'f(' wilh Ih!'
relalionship

undt'r sludy.

I>ls('overs and ex poses causal relation.

ships under conlrolled condltlons.
Slalemcnls of Rrealer riRor are Olade
po~slhle and Increasl.-d validlty of social
If('alments or pror.1ms Is
demon~lrall.'(l.

Indl.pt'ndf'nl vari.lhll' (lrt'alml'nl) 15

Cilpable of wlde varlallon; d sensltiw
or dcfinltive crilerlon variable Is found.
M.tlehln dala provldl' the strnnRI.'51
ponlhll! control.
A I:ausal pallt~rn may bt. Infermd wlth
hih cnnfldenee.

-+ - 0---

22

_+

oo_

0______.._

HANOBOOK OF RESEARCH DESIGN AND SOCIAl. MEASlJREMENT

References
Bales, Robert. (1969). /'crsonality a"d inter!,erscmal behavior. New York: Holt, Rineharr, ami
Winston.
nlau, Petcr M., & Duncan, O. D. (1967). The American occu!,aticmal structure. New York: Wiley.
I\urgess, E. '.1v.,& Cottrcll, I.conard S., Jr. (19')8). /'redicti"g SIIccessor (ai/urc'in marriagc'. I.ondon:
Routlcdgcffhocmmes Press.
Chapin, F. S. (1974). Ex!,erimental designs i" soci%gical research (Rev. ed.). Wcstporr, CT: Grecnwood.
Comell, Frcd '.1v.(1971). The Railmader. New York: Irvington.
I>avis,Jamcs A. (1965). Undergraduate career decisicms: Corre/ates o( occu!,aticmal choice. Chicago:
Aldinc.
I>uncan, GregJ., & Morgan,James N. (Eds.). (1980). Five thousand American (ami/ies: Vol. H.Analyses o( the (irst eleven years o( the Panel Study u( /ncome f)ynamics. Ann Arbor, MI: Survcy Rcsearch Center, Institute for Social Research, Univcrsity of Michigan.
Eddy, Elizabeth. (1969). Becoming a teacher: The passage to pm(essional status. Ncw York: Tcachcrs
College Press.
Glueck, Sheldon, & Glueck, Elcanor. (t 959). Predicting delinquency and crime. Cambridgc, MA:
Harvard University Prcss.
Hammond, Phillip E. (Ed.). (t 964). Socio/ogists at work: Essays cm the cra(t o( social research. Ncw
York: Basic Books.
Hollingshead, August B. (t 948). Elmtown's youth: The impact o( social c/asses on adolescents. Ncw
York: Wiley.
Hollingshead, August B. (1975). Elmtown's youth and Elmtown revisited. New York: Wiley.
Horst, Paul. (1941). The prediction o( personal ad;ustment: A survey o( logical problems and research
techniques, with i//ustrative application to problems o( vocational selection, school success, marriage, and crime. New York: Social Scicnce Rcsearch Council.
Kinsey, Alfred c., & Associates. (t 998). Sexual behavior in the human (emale. Bloomington: Indiana
University Press.
Kinsey, Alfred C., Pomeroy, Wardell B., & Martin, Clydc E. (t 998). Sexual behavior in the human
male. Bloomington: Indiana University Press.
Lynd, Robert, & Lynd, Helen. (1975). Middletown: A study in American culture. New York: Harcourt, Brace.
Lynd, Robert, & Lynd, Helen. (t 982). Middletown in transition:A study o( cultural con(licts. New
York: Harcourt Brace Jovanovich.
Miller, James Grier. (Ed.). (1950). Experiments in social process: A symposium on social psycho/ogy.
New York: McGraw-Hill.
Sewell, William H., & Hauser, Robert M. (t 975). Education, occupation, and earnings: Achievement
in the early career. New York: Academic Press.
Smigel, Erwin O. (1969). The Wa// Street lawyer, a pro(essional organization man? Bloomingron: Indiana University Press.
U.S. Department of Commerce, Bureau of the Census. (various years). Current population survey.
Washington, DC: Author.
Whyte, William F. (1993). Street corner society: The social structure o( an Italian slum. Chicago: University of Chicago Press.

Further Readings on Research Design and Selecting a Research Problem

AIlwood,Carl Martin, & Barmark,Jan. (t 999). The role of research problems in the processof research. Social Epistemology, 13,59-83.

PART 2: BASIC RESEARCH DESIGN

~ 23

Bailey, Kenneth D. (1999). Methods of social research. Collingdale, PA: Diane.

Bechofer, Frank, & Paterson, Lindsay. (2000). Principies of research design in social sciences. New
York: Routledge.
Garratt, Dean, & Hodkinson, Phil. (1998). Can there be criteria for selecting research criteria? A
hermeneutical analysis of an inescapable dilemma. Qualitatille lnquiry, 4, 515-539.
Grunow, Dieter. (1995). The research design in organization studies: Problems and prospects. Organization Science, 6, 93-103.
Hazen, Dan, Horrell, Jeffrey, & Merrill-Oldham, Jan. (1998). Selecting research col/ections for
digitization. Washington, DC: Council on Library and Information Resources.
Mitchell, Mark, & Jolley, Janina. (1995). Research design explained. Fort Worth: Harcourt College
Publishers.

Reiser,StanleyJoel, & Bulger,Ruth Ellen. (1997). The social responsibilitiesof biologicalscientists.

Science and Engineering Ethics, 3, 137-143.
Sasaki, Masamichi. (1995). Research design of cross-national attitudinal surveys. Behalliormetrika,
22,99-114.
Tabachniek, Barbara G., & Fidell, Linda S. (2000). Computer-assisted research design and analysis.
Needham Hc:ights, MA: Allyn and Bacon.

HOW 5CIENCE15BUILT

This guide sets forth the canons of science as seen by the behavioral researcher and should be used as a
ser of signposts to poinr rhe direcrion of and idenrify rhe possible diffieulties inherent in any scientific
exploration. It's likc a digcst of the philosophy rhar researchers carry with them, use in their work,
and live by amid rhe ups and downs of a life conducting research.
The three brief essays rhat follow describe (a) the importance of conceptual definitions and theory
formularion in the construction of scientifie knowledge (DiRenzo), (b) assumptions underlying the
application of rhe scientifie method (Sjoberg and Nett), and (c) dilemmas of the researcher and the
disrinctiveness of behavioral science (Kaplan). They are all well worth reading as you begin your
thinking about research topies and professional activities you would like to pursue.

~ Importance

of Conceptual

Definltlon and Theory Formulatlon

4()nl()n J. l>iRenz.o -Scientificinvestigation seeks to explain the phenomena it studies inour world of experienee;
by establishing general principies withwhichto explain them, hopefully,scienee can predict sueh
phenomena. The principies of scienee are stated ultimately inwhat are known as theories. Toexplain the facts of reality,scientists require an organized system of concepts. A "scienee without
concepts" is an impossibility-as unthinkable as any form of rational activitywithout eoncepts
would be. Yet,to say that coneepts are indispensable to science is merely to presuppose or to
make possible the problems, namely, the definition and formation of the required scientific elements.
Initially,inscientificinquiry,description of phenomena may be stated ina nonteehnieal vocabulary.The growth of a discipline soon involvesthe development of a system of speculation, more
or less abstraet, of concepts and corresponding terminology. Nevertheless, even after decades
of definition, and redefinition, many of the fundamental terms in the sciences are far from belng