QUALITATIVE METH PH
QUALITATIVE METH PH HSERV 526
Popular in Course
Popular in Health and Human Services
This 13 page Class Notes was uploaded by Lincoln Hintz III on Wednesday September 9, 2015. The Class Notes belongs to HSERV 526 at University of Washington taught by Stephen Bezruchka in Fall. Since its upload, it has received 20 views. For similar materials see /class/192010/hserv-526-university-of-washington in Health and Human Services at University of Washington.
Reviews for QUALITATIVE METH PH
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/09/15
HSFRV 5261331 526 Qualitative Research Methods for Public Health Winter 1998 University of Washington Seattle WA 98195 Course Instructor Steve Bezruchka MD Lecture 10 Indirect Observation and Group Process Learning objectives 7 descnbe same unublm v2 mea ubservalmn new Wuvk r CnmmenIsQueslinns lrnm prwinus Sessi r Exerc39se 5V turn in me I39slV pilesn exercise 5 discuss Grnup prncvss Pvugvam piarmrrrg program a prearranged set or acwmes Wmch specrry the means to acmevet process underiyrng deveiopmem or modrrrcauon or programs 7 k2 rncvssvs lt 3 pvehmmavyvewew A d25ignampimpemenlalmn elmgs srtuatrun Wham membevs utgruup agvee upun deswed gua ampl2chnuugies exisHu chieve this gua ucus un enurematmn amp mturmatrun exchange 7 leadevrcemeved r negutratmn m havgammg r uppusmg taetruns emermg m nuvms vaiues uvvesled rmereas uppusmun in each ulhev cuncemmg ermer ends means m bum stand in are mere ways in deveiup eunsen Idea generation Group Process vs Individual Process Individual ldea Generation general impression group participation facilitates idea production person can think up twice as many ideas in a group than alone number of ideas produced by a individuals independently is considerably less than produced by an equal number of individuals together studies confirm this organizational life can be endless meetings with little productivity administratorsplanners have little time to work alone Group ldea Generation Brainstorming shared problem solving all members of a group spontaneously contribute ideas group settings can hinder creative process intimidate people very culture dependent First Nations Asians others not wanting to confront or be critical or be ridiculed physical arrangement of space principles for generating ideas have clearly defined subject try to maintain criticism free setting when ideas generated so ideas flow freely not be hampered by critical comments or attacks from other group mem e s Bill Gates that s the stupidest thing I have ever heard all ideas receive approval or neither praised nor criticized related atmosphere sufficient time small group leader or not Limitations discussion generally succumbs to influence of few individuals due to status personality amp other forces Post idea generation analysis is separate function apart from idea generation ranking of ideas by group mem task force to review suggestions amp make recommendations group leader to choose Nominal Group Technique NGT nominal refers to process in which individuals are together in a group in name only since verbal exchange is excluded or limited re ects silent and independent generation of ideas History developed and tested in late 1960 s by Van de Ven amp Delbecq derived from studies of decision conferences of aggregating group judgements and from social work studies of problems surrounding citizen participation in program planning Objectives assure different processes for each phase of creativity in problem solving fact finding phase problem search amp generation of data about problem or about different proposed solutions evaluation phase information synthesis screening and choosing among strategic elements of a problem or component elements of alternative solutions balance participation among members voting techniques for group judgment Process Preparation members members can be heterogeneous unlike a focus group heterogeneous groups can be more creativebut there can be communication difficulties individuals should be able to speak a common language if have doctors and patients doctors should not use medical jargon could have a variety of social scientists members in an NGT should be interested in the problem and have either experience or education which makes them a resource to the flexible so they can openly explore various points of view selection bias if suspect try a different mixture of participants in future groups leader skills necessary understanding of NGT process selfconfidence to lead group through process steps legitimate so accepted by grou female nurse poor leader for male research medical staff ensure all participants are able to express their views and the keep particular personal or professional views from dominating the discussion supplies table amp chairs flip chart for each table 3x5 cards for each table felt pens paper amp pencil for each participant focus of NGT meeting is on list of ideas placed on flip chart rather than on individual participants so open U with flip chart at open end of table 5 to 10 participants leader or recorder determine purpose of meeting structure proposed question assemble members leader s welcoming statement should include cordial welcome sense of importance concerning the group s task members gathered because of analytic abilities amp problem solving skills clarification of the importance of each member s contribution indication of use or purpose ofthe meeting s output leader serves as recorder resists nonprocess clarifications has question in writing PRETEST THE QUESTION models good group behavior by writing in silence sanctions individuals who disrupt the silent independent activity avoids leading any answer that comes to your mind should be written on the worksheet 1 silent generation of ideas in writing adequate time for thinking facilitates work by other group members reflecting amp writing together ie constructive tension created by observing other group members working hard avoids interruption of each other s thinking sufficient time for search amp recall avoids premature focussing on single ideas eliminates dominance by highstatus or aggressive members in idea generation keeps group problem centered N roundrobin feedback from members to record each idea one idea from each group member in each go around written on the paper or blackboard in a terse phrase on ip chart equalizes opportunity to present ideas only one idea each time around members can hide substantial number of ideas if had to present entire list at once separate ideas from personalities since who contributed what is not stressed with each persons ideas are not clustered provides written record and guide ideas to be presented in brief words or phrases variations on themes are desirable can give numbers or letters as a label increases ability to deal with large number of ideas avoids loss of ideas confronts group with an array of clues encourages hitchhiking ideas written on flip chart may stimulate another member to think of an idea he had not written before so he can add that idea amp report it places conflicting ideas briefly and comfortably in front of group forces group to fully explore problem write ideas on language used by participant group s patience on this step is brief so write as quickly as possible 3 serial discussion of each recorded idea for clarification and exploration Guidelines leader verbally defines the role of the step as clarification leader should pace the group to avoid arguments or neglect of some items at expense of others avoids discussion focussing on particular idea helps eliminate misunderstanding opportunity to express logic behind items allows members to disagree without argumentation 4 individual voting on priority ideas independent discrete judgments in writing helps eliminate social pressures can rankfrom 5 to most important item to 1 for least important could place this on separate 3x5 cards collected shuffled amp recorded can also have numbersletters on each of 5 cards amp rank then in order for those nonliterates five items are about the most you can rank expressing judgments by rankordering increases accuracy of consensus displaying the array of individual votes highlights areas needing further clarification or discussion 5 discussion of preliminary vote shows the distribution of the group s response provides group members with final opportunity to clarify their positions ensures that spread votes really reflected differences of judgment not unequal information or misunderstanding people can change their minds voting discussion then revoting a more accurate indication of preferences than voting alone 6 Final vote can rank as before or rate each item Advantages different group processes for independent idea generation feedback amp voting increased attention to each idea amp increased opportunity for each individual to assure his or her ideas are heard avoids arguments amp extraneous interactions more egalitarian ensures participation avoids dropping out may generate more options amp ideas per unit time 2 people arguing takes up a lot oftime keeps discussion focussed Limitations crossfertilization of ideas diminished because of structure imposed by NGT reduced flexibility structure is imposed on meeting puts people on the spot to produce somethin composition of group is important need minimal level of education bringing people together may be costprohibitive but consider the use of videoconferencing Comparison with Brainstorming average number of unique ideas average total number of ideas quality of ideas produced varies can have collective ignorance rather than wisdom triangulate research suggests that when group task is to generate information on a problem idea generation interacting groups inhibit creative thinking Delphi Technique after exercise Table 22 Comparison of Qualitative Differences among Interacting NGT amp Delphi Groups Delbecq A A pg 32 Class Exerci Question TABLE 2 2 Cmupznwn of Qualiiaiivc Di crcntc Aiming Inrerading NGT mi Delphi Groups nurLe aimed imm amp DeciriowMakln E eciwencxs by ndrcw Vun dc Van published by enler for Busmc si and ECkann lC Resend Pros Ken an Univv ily l974 Uu d by permission EIMENSION39 39 7NTERACriNG NGT 7quot W DELPHI A 0v ALI Unstrucliiredmm39linu Smicrurud mac n i i ml lt M Tl lODOLOCV High mmilyilm39 littween tilWin making groups ROI ORIENTATION OF GROUPS llFlA iv ifQLiANTi39iY OF ll A S Social viiiuimnal lmu low locum quotmiquot offer ii g rm iwiahiliiv between dc cisionrmnking groups llalanrul mcml cmminnal imi mi insnuiiicmzl mm iimii imlupunilunl ihinli ng RH ATlV QUALlTY AND si i 39 BTURMNI IV Vioui Liiw quaim k nvmlIiiiiunx ll llll ll umluvmilv pr um iii twain lligli mm n rut m lnr mmniunuiu of quma mm mi illmltk xcpnns Aw miiiiiiuv liclwccn icci Trxlt imminwnml rams High isiianI thinking llmli wim I igi mist I r i Inn nui lu mnfurm 912mm ii XVIOR EQUAL Y PARTICIPATION METHODS OF INl ll RFQOLU39I 0N quotm DEL lSION MOTIVATION Wh N om inal Group Technique Break into two groups of 78 each be recor e select one silently generate ideas in urn wri e eac 39d l llVl Si iwrnhluiii rutm TN nmilnncc mimy m knnwlLdge Nuv y Mimiquot luillln39 l ruaciiw mmlrd pmlwluni ixm l lligli ixxL nonicrmlncs New mi imi ml kmi unligc Mm m min 7 V k39xpiymlinl 39 ii i l l mauwc 39 iiumlloi ligi New wk lnuwiugc mm fun 5 Ciunlily in pmi in nrcpcndcnl judx intm l tmin wnirn Smoothing mu quotmi uilidmwnl Lad ml iiosum L v full atcmiiplisl iiiwm Mmliiim in wri in mum Lunmrud umrmnmm aml Wantm mixing lligli closum llinli lull zunmplishmem ligh present the question in writing I t g a on blackboard or ip chart without debate imiimm onward Majan rquotin of pooled inde pendent iudgmnms lligli Lilquotc Medium sci Accomplidimcm Mciilum se Comparing NOMINAL GROUP PROCESS with Brainstorming at should be done to lessen the stress on public health students at UW move group discussion serially through ideas for clari cation and elaboration but not advocacy and display preliminary vote on item importance in descending order hold briefdiscussion for clari cation interpretation and re nement of results obtain analyze and display nal vote as descending order of ranked priorities Review of Nominal Group Technique question should threatening than other forms ofridea generation ifuse ip chart ha ve a recor d ofthe me eting preferable to blackboard be pretested and related to the objective ofthe meeting leader should avoid leading the group with exemplary an focus groups can clarify issues and help frame the challenge question for the NGT 39 quot 39 39 from quot quot 39 backgrounds less round robin idea generation equalizes opportunity to present ideas only one idea each time aroun members can hide substantial number of ideas if had to present entire list at once separate ideas from personalities since who contributed what is not stressed with each persons ideas are not clustered discussion of ideas canshould proceed serialy to avoid focussing on one and clarifying used with minimal literated and with aphasic people stress that worksheets will not be collected recorder has to assist participants with what is written on flip charts Applications developing treatment guidelines by providers for various diseases improving efficiency in pharmacies input from manager pharmcist fulltime and parttime sales people generates input from all employees improving curriculums and solving other problems in various educational institutions used on students amp administrators use of learning contracts among nursing students implementing different models of nursing practice in South Dakota developing university student safety programs understanding adolescent perceptions affecting desire to seek care eg provider hand washing turns out to be very important caregivers in chronic care settings communicating with residents having different disabilities eg dementia aphasia little change in communicating with people having such disabilities patients in nursing homes evaluating different measures of functional status Appropriate for identifying elements of a problem situation identifying elements of a solution program establish priorities where judgments of several individuals must be aggregated into a group decision CONSENSUS BUILDING Table 1 Relative Strengths of NGT brainstorming Delphi and PG Family Practice 10 7681 1993 mm mmmnvmmwihxqNm lummlornmm Dl39llllllIlrlurw ynmly Advunlalgw NOT Brillnslurlmllg mlle Dllllcull l39nr dunlillanl m No U pzlrliClpanh ll L39ulllrul Av qule Ye No ya deuwn lllllking Licrlcnllcs a lllgll Vcs Possibly m numlmr l l39 L39Ummclllildclh Provides mppml Iu Ya No Nu allow idcmlflmllml 0 personal problems and xull clnslln Allows mummncm Yes No vm m llupomncc ul ldcth ilclm lo lmlwullmn Avold pllrsull ul u Yin Nu Yc illglc Ham 0 lllmlglll rimmo cm l cmlmgcs lllilmrily Ya No Yes concenlxoplmm m m vmccd gt niuipam value ruminy M No s l uemclmu Liv gmup cuhcslvcllcsx High degree of m Yes No Vt complclinll Euw ul adlnlllllll alinn No v Yes Yes No No Need for cxpcrmwl leader Delphi Technique a series of questionnaires first used in 1948 currently used by NIH CDC etc Objectives to develop a range of possible program alternatives to explore underlying assumptions or information leading to differentjudgments to seek out information which may generate consensus on part of respondent group to correlate informed judgm nts on a topic spann39n a wide range of disci lines to educate the respondent group as to the diverse amp interrelated aspects ofthe topic Process rst questionnaire asks individuals to respond to a broad question could focus on problems objectives solutions or forecasts each subsequent questionnaire is built upon responses to preceding questionnaire quot quot J by 39 39 have enough information exchange Groups ofind39 39duals necessary decision makers expect some sort ofproduct from exercise staff group designs initial questionnaire summarizes returns redesigns followup questionnaires respondent group lucm l msllvly Pmny I lmlbly l omllly l mxil y I umbly l msibly Vex those whose judgments are being sought who are asked to respond to questionnaires Issues takes a long time is respondent group anonymous use openended or structured questions how many iterations of questionnaires amp feedback reports what decision rules used to aggregate judgments of respondent group Advantages views expressed impersonally Limitations panelist fatigue after 23 rounds Developing Consensus consensus methods try to determine extent to which people agree on a given issue need to have solvable problems need to balance knowledge ampjudgement level or type of consensus must be defined in advance avoid groups where domination by an individual or coalition representing vested interests both Delphi and NGT allow relative or complete anonymity processes occur in rounds allowing individuals to change their opinions have controlled feedback showing the response after the vote in NGT have group response numerically rather than a statement with no sense of dissention agreement has two forms extent to which each respondent agrees with issue under consideration extend to which respondents agree with each other INDIRECT OBSERVATION Study people s behaviors without their knowing it is less reactive people don t respond and behave differently because they are being observed called unobtrusive observation by Bernard There are ethical dif culties here Hand washing in public bathrooms NYT 960917 60 in Penn Station washed up to 78 in chicago phone survey 94 said they washed their hands Physical clues which indicate what a person s behavior was behavior trace studies eg looking at wear spots on a floor or carpet to see where people went graffiti NIH bathrooms doodles on class notes garbage garbology a trace measure Rath39e at U of Arizona has been studying behavior patterns by analyzing garbage from a representative sample of residents since 1973 Bernard Pg 334 for details more accurate than reported behaviors many studies confirm this Marker behaviors an easily observedless sensitive behavior used as an indicator of another difficult to observesensitive behavior e g defecation or sexual behavior someone in Nepal carrying a Iota is going to defecate sexual behavior of a population is very difficult to study and this is faced in current AIDS epidemic study of bathroom graffiti in Manila and Chicago 42 in Chicago dealt with homosexuality compared with 2 in Manila Archival Research Kroeber s study of women s fashions including measurements ofwaist diameter skirt lengths etc from fashion journals Found a cycle of skirt widths over a full century and a skirt length period ofa third ofa century Bernard 337 records of births deaths causes of death etc eg L A County had 82 homicides in 1953 and 2512 in 1992 while population doubled incidence increased almost 15 times or 1431 Retail Anthropology New Yorker article on The Science of Shopping Nov 4 1996 15 Other examples in Bernard Chapter Difficulties with lndirect Observations people may change their behavior when they learn you are doing this in garbage project people put out fewer liquor bottles very time consuming expensive how far do people go to get data Techniques of Deception Bathroom Hand Washing Study reported in NYT 960917 60 washed hands in Penn Station public bathroom up to 78 in Chicago while phone survey 94 said they washed their hands Tearoom Trade Study an example of disguised observation and can the data be protected Bernard pg 348 on fleeting homosexual behavior in public washrooms Laud Humphreys observed homosexual acts while acting as a lookout He recorded identification data on subjects and five years later interviewed them while not revealing his identity Would a researcher go to jail to protect data that could be incriminating ethnical issues discussed in Bernard Chapter 15 Example Intrahousehold Food Allocation in Rural Nepal Gittelsohn J 1991 Opening the box intrahousehold food allocation in rural Nepal Soc Sci Med 3310 1141 54 Background differential nutritional status malnourished children found in households where siblings approach normal ht wt food distribution reflects order of precedence and social value of food consumers differences in morbidity amp mortality within households interventions programs directed at households may not consistently reach targeted indiVIduals intrahousehold food intake and distribution studies have not looked at behaviors during meals and this influence on intake black box preferential food distribution increased quantity or quality patterns sex male adults favoring male wage earners because of economic contribution to hh favoring children based on perceived future contribution to hh human capital body typesize tied into cultural perceptions of ideal hypotheses for Nepa preferential distribution to males of all ages and senior members overjunior preferential distribution as certain foods inappropriate for infants children lactating women pregnant women sick etc Food serving elements regarding allocation identity of food server order in which individuals served what is served to w om how much is served to whom how individuals are served Food allocation patterns egalitarian persistent or permanent favoritism in food quality or quantity temporary shifts in allocation due to transitory states pregnancy illness Methods levels of information community focus groups key informant interviews archival sources household environment sanitary conditions economic status weekly food frequencies decision making foodways food classification systems infant feeding practices case histories unstructured direct observations of household behaviors individuals n767 demographic weekly morbidity recalls anthropometry time allocation 24 hr dietary recalls direct structured observation of meals in 105 hh randomly selected from 6 study communities comprising 030 of hh in each communi types amp quantities of foods consumed by hh members during meals associated behaviors total of 318 complete meal observations kharchha or rat39ko kharchha each day study community selected randomly observers randomly assigned to hh amp continued to observe that hh throughout study hh not notified in advance that their meal was to be observed Data recorded activities conversation food food condition food quantity trained to estimate quantities by eye dietary recall at end of meal observation for previous meal Analysis food events Table 1 serving order Table 2 small children of both sexes have top priority in serving order after 10 years male serving order scores remain fairly constant while female scores decrease with age then female scores increased with old age reflecting increased status of elderly women in little difference between males amp females up to 10 years of age serving method Table 3 little variation in serving method by sex up to age 18 automatic serving of males with females serving themselves is adult practice jutho concept with women serving men second helpings Table 4 ages 318 amp for elderly males receive higher second helping preference than females but statistically significant only for 710 amp1518 yr age groups adult women receive second helpings more frequently than men perhaps because they serve themselves frequently in small portions also eat leftovers refusals Table 5 few amp non discriminatory channeling food offered to one person not to another Table 6 children channeled more foods than adults no sex preference until age 10 amp significant in early and midadulthood food quantity Table 7 decreases with increasing age no sex differences among children but males get more than females over age correlations between mechanisms of preferential food distribution Table 8 higher priority in serving to adult males and children but they get second helpings less frequently amp may be served a larger initial portion those who serve themselves adult women have lower serving priority lower channeling and food quantity but more second helpings food proscriptions and prescriptions foods not served to adult women soybeans wild green leafy vegetables potato pickle banana mango fish eggplant cow milk yogurt cow milk ghee buffalo milk lassi chili wheat purl wheat rot39with oil jamuna pork chicken eggs amp liquor above considered difficult to digest by infants amp avoided by nursing women key informant interviews support notion that food belief systems are reason they aren t eaten animal products in great demand and distributed to adult males amp small children who request them little snacking by food preparers as it would pollute food utho game NUTRITIONAL CONSEQUENCES mean intakes by kcals Table 9 adequate for children age 1 to 9 children favored insufficient for high activity levels of girls amp young women or for needs during pregnancy amp lactation low for young adult males 1825 mean intakes of betacarotene Table 10 children exceed minimum because milk channeled only to them adult women have low intake Results dietary adequacy of children appears better than of adults with sex differences in quality pronounced only in adulthood adult males received preferential food allocation served after small children automatically served received second helpings of food automatically favoritism in terms of substitution amp channeling disproportionately larger shares of animal products adult women disfavored eat last decreased access to second helpings limited quantities except for pregnancy lactation never received channeled foods eat disproportionately smaller amounts of more costly amp rare foods feeding mechanisms daughtersinlaw had low status do most of heavy domestic work served last or next to last even if not the food server rarely asked for second helpings amp worked under critical eye of motherinlaw food allocation disfavor adolescent girls adolescent boys served automatically offered second helpings channeled etc elderly have high status both sexes no difference in treatment of small children under age 8 by sex Preferences senior adults favored overjunior adults and children over adults No neglect of female children seen in other South Asian settings gifts from bride s hh have less value than dowries in N India or Tarai amp items remain property of bride after marria e girl children are highly productive in hh child care fetch water amp wood clean process food Program Implications efforts to improve children s nutrition in this region should be directed at both sexes focused on interaction of food server amp child encourage mothers to feed children who do not request food as well as those that d o young pregnant amp lactating women most at risk target educational messages at food server Questions what determines dietary intake food serving behaviors or individual food preferences amp how does this vary by caste economic status education how to influence policy makers how to identify lessfavored individuals within hh amp direct resources to them Required Reading Bernard 332336 McMurray AR Three Decisionmaking Aids Brainstorming Nominal Group and Delphi Technique Jr of Nursing Staff Development 1994 102 p 6265 Recommended Reading Jones J et al Consensus methods for medical and health services research BMJ 1995 3117001 37680 Pelto amp Pelto 103112 Delbecq A A Van de Ven and D Gustafson Group Techniques for Program Planning A Guide to Nominal Group and Delphi 1986 Middleton WI Green Briar Press Lect 10doc 73010
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'