accessing hard to reach populations for research.doc

Issue 33 , Summer 2001 Social Research Update is published quarterly by the Department of Sociology, University of Surrey, Guildford GU7 5XH, England. Subscriptions for the hardcopy version are free to researchers with addresses in the UK. Apply to SRU subscriptions at the address above, or email Accessing Hidden and Hard-to- Reach Populations: Snowball Research Strategies Rowland Atkinson and John Flint Rowland Atkinson and John Flint are
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
  Issue 33 , Summer 2001 Social Research Update is published quarterly by the Department of Sociology, Universityof Surrey, Guildford GU7 5X, !ngland Subscriptions for the hardcopy version are free toresearchers #ith addresses in the U$ %pply to SRU subscriptions at the address above, or email sru&soc surrey ac u'  Accessing Hidden and Hard-to-Reach Populations: Snowball Research Strategies Rowland Atinson and !ohn lint   Rowland Atinson and !ohn lint areresearchers at the #epartment o$ %rbanStudies& %ni'ersit( o$ )lasgow* +oth ha'ean interest in the spatial distribution ande,perience o$ social e,clusion and ha'ebeen commissioned to de'ise amethodolog( $or tracing residents wholea'e regeneration areas in Scotland*  In its simplest $ormulation snowball sampling consists o$identi$(ing respondents who are then used to re$er researchers on to other respondents*  Snowball sampling contradicts man( o$ the assumptions underpinning con'entional notions o$ sampling but has a number o$ ad'antages $or sampling populations such as the depri'ed& the sociall( stigmatised and elites*  Snowball sampling has ad'anced as a techniue and the literature contains e'idence o$ a trend toward more sophisticated methods o$ sampling $rame and error estimation*  Although the( 'iolate the principles o$ sampling& the use o$ snowball strategies pro'ides a means o$ accessing 'ulnerable and more impenetrable social groupings*  Howe'er& the nature o$ similarit( within social networs ma( mean that .isolates/ are ignored* (reading an uneasy line bet#een the dictates of replicable and representative researchdesign and the more flo#ing and theoretically led sampling techniques of qualitativeresearch, sno#ball sampling lies some#hat at the margins of research practice o#ever,the technique offers real benefits for studies #hich see' to access difficult to reach or hidden populations (hese are often obscured from the vie# of social researchers and policy ma'ers #ho are 'een to obtain evidence of the e)periences of some of the moremarginal e)cluded groups *olicy ma'ers and academics have long been a#are that certain +hidden populations,such as the young, male and unemployed, are often hard to locate -ther groups such ascriminals, prostitutes, drug users and people #ith unusual or stigmatised conditions .e g %/DS sufferers0 pose a range of methodological challenges if #e are to understand moreabout their lives (his issue describes the processes and the advantages and difficulties of using sno#ball sampling techniques Snowball Sampling Sno#ball sampling may simply be defined as1 % technique for finding research sub2ects -ne sub2ect gives the researcher the name of another sub2ect, #ho in turn provides the name of a third, and so on .3ogt, 40 (his strategy can be vie#ed as a response to overcoming the problems associated #ith sampling concealed populations such as the criminal and the isolated .6augier and Sargeant, 470 Sno#ball sampling can be placed #ithin a #ider set of lin'tracing methodologies .Spreen, 480 #hich see' to ta'e advantage of the social net#or's of identified respondents to provide a researcher #ith an evere)panding set of potential contacts .(homson, 470 (his process is based on the assumption that a +bond or +lin' e)ists bet#een the initial sample and others in the same target population, allo#ing a series of referrals to be made #ithin a circle of acquaintance .9erg, 4::0 Sno#ball sampling can be applied for t#o primary purposes 6irstly, and most easily, asan +informal method to reach a target population /f the aim of a study is primarilye)plorative, qualitative and descriptive, then sno#ball sampling offers practicaladvantages .endric's, 9lan'en and %driaans, 480 Sno#ball sampling is used mostfrequently to conduct qualitative research, primarily through intervie#s Secondly,sno#ball sampling may be applied as a more formal methodology for ma'ing inferencesabout a population of individuals #ho have been difficult to enumerate through the use of descending methods such as household surveys .Sni2ders, 48; 6augier and Sergeant,470   !arly e)amples of the technique can be seen in <hytes Street =orner Society .4550 and*atric's study of a Glasgo# gang .47>0, #hich used initial contacts to generate conte)tsand encounters that they could use to study the gang dynamic % general move a#ay from participant observation of this 'ind to#ards the use of sno#ball sampling techniques primarily for intervie#based research has been seen more recently Sno#ball samplinghas been used in studies of drug users .%vico et al, 4:, Griffiths et al, 4>; $aplan etal, 4:70; prostitution .?c@amara, 4A0; pic'poc'ets ./nciardi, 4770; aids sufferers.*olla' and SchlitB, 4::0 and the seriously ill .Sudman and 6reeman, 4::0 <hile some may see' to characterise the topics for #hich sno#ball strategies have beenused as being trivial or obscure, the main value of sno#ball sampling is as a method for obtaining respondents #here they are fe# in number or #here some degree of trust isrequired to initiate contact Under these circumstances, techniques of +chain referral mayimbue the researcher #ith characteristics associated #ith being an insider or groupmember and this can aid entry to settings #here conventional approaches it find difficultto succeed % range of advantages have been claimed for sno#ball sampling 6irstly, it has enabledaccess to previously hidden populations -ften members of such populations may beinvolved in activities that are considered deviant, such as drug ta'ing, or they may bevulnerable, such as the stigmatised in society, ma'ing them reluctant to ta'e part in moreformalised studies using traditional research methods (rust may be developed asreferrals are made by acquaintances or peers rather than other more formal methods of identification Sno#ball sampling has been found to be economical, efficient andeffective in various studies 6or e)ample, it has been sho#n to be capable of producinginternationally comparable data in %vico et als 4:: study of cocaine users in three!uropean cities /t may be used to e)amine changes over time Sno#ball sampling canalso produce indepth results and can produce these relatively quic'ly *erhaps one of the strongest recommendations for the sno#ball strategy stems from adistinction bet#een descending and ascending methodologies .3an ?eter, 4C0 (raditional techniques such as household surveys, as +descending strategies, areassociated #ith a largely quantitative tradition of the measurement of social problems thatoften suffer from a lac' of responses from particular groups, often the young andunemployed males %scending methodologies, such as the use of sno#ball techniques,can be used to #or' up#ards and locate those on the ground #ho are needed to fill in thegaps in our 'no#ledge on a variety of social conte)ts /n this sense sno#ball samplingcan be considered as an alternative or as a complementary strategy for attaining morecomprehensive data on a particular research question <hile many have considered sno#ball strategies primarily as an aid to accessing thevulnerable or the deviant, some studies have used them to engage #ith the +hard to reachamong urban elites Saunders .47>0 study of urban politics is an e)ample #here a+reputational method #as used Respondents #ere as'ed #ho held po#er in the localarena (his led to a series of contacts and the establishment of a sub2ective indication of   the relative local po#er bases (his suggests that sno#ball sampling has a #ider applicability in sociological research than has hitherto been realised #i$$iculties Sno#ball samples have a number of deficiencies 4 *roblems of representativeness and sampling principles (he quality of the data and in particular a selection bias #hich limits thevalidity of the sample are the primary concerns of recent sno#ball samplingresearch .3an ?eter, 4C; $aplan et al, 4:70 9ecause elements are notrandomly dra#n, but are dependent on the sub2ective choices of therespondents first accessed, most sno#ball samples are biased and do nottherefore allo# researchers to ma'e claims to generality from a particular sample .Griffiths et al, 4>0 Secondly, sno#ball samples #ill be biasedto#ards the inclusion of individuals #ith interrelationships, and therefore#ill overemphasise cohesiveness in social net#or's .Griffiths et al, 4>0and #ill miss +isolates #ho are not connected to any net#or' that theresearcher has tapped into .3an ?eter, 4C0 (he problem of selection bias may be partially addressed, firstly through thegeneration of a large sample and secondly by the replication of results tostrengthen any generalisations %t present, a statistical formalisation of sno#ball sample biases is not available .3an ?eter, 4C0 o#ever, larger sample siBes may reduce bias 6or e)ample *olla' and SchlitBs study of %/DS sufferers produced a sample #ith representative proportions for age,class and siBe of to#n of residence .*olla' and SchlitB, 4::0 /n addition to selection bias, there is also the issue of gate'eeper bias .seeGroger, ?ayberry and Stra'er, 40 /n their #or' they identified a difficulty#hen using nursing home staff as +gobet#eens in obtaining the informedconsent of caregivers (hese +gate'eepers #ere sometimes reticent or  protective to#ard those they cared for and sometimes hindered access for theresearchers 9ased on their e)periences they attempt to dra# a comparison bet#een sno#ball sampling and scrounging sampling (hey describe the latter in terms of1 desperate and continuing efforts, against mounting odds, to round out the collection of individuals #ith relevant types of e)periences #e 'no# to e)ist but have not been able to capture .Groger et al, 41:>C0<hile social scientists may vary in the degree to #hich they accept such a vie#point #e can recognise similar traits in quantitative approaches such as household surveys ere #e often find struggles to obtain adequate numbers
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!