Xã hội học - Selecting samples and negotiating access

Selecting Samples and Negotiating AccessSamuel K. Frimpong (Ph.D.)Outline presentationThe need to sampleOverview of sampling techniquesDeciding on a suitable sample sizeChoosing the appropriate sampling techniqueChecking the sample representativenessStrategies to gain accessWhat is Sampling?Sampling refers to the techniques that enable you to reduce the amount of data you need to collect by considering only data from a subgroup rather than all possible cases or elements. Population SampleSampling vs. CensusA census involves collecting data from all members of a population.A sample is a relatively small subset of the population that is selected to be representative of the population’s characteristics.The Need to SampleIt would be impracticable for you to survey the entire populationYour budget constraints prevent you from surveying the entire populationYour time constraints prevent you from surveying the entire populationYou have collected all the data but need the results quickly Sampling Design ProcessThe sampling design process involves answering three questions:Should a sample or a census be used?If a sample, then which sampling approach is best?How large a sample is necessary?Steps to obtain representative sample1. Define the target population. 2. Choose the sampling frame. 3. Select the sampling method. 4. Determine the sample size.5. Implement the sampling plan.Defining Target population and Sampling FrameTarget population is the complete group of objects or elements relevant to the research project. They are relevant because they possess the information the research project is designed to collect.Sampling frame is a complete list of all the elements in the population from which the sample is drawnSampling Techniques/MethodsProbability or representative sampling: each element of the population has a known, but not necessarily equal, probability of being selected in a sample.Non-probability or judgmental sampling: not every element of the target population has a chance of being selected because the inclusion or exclusion of elements in a sample is left to the discretion of the researcher.Examples of Sampling methodsProbabilityNon-ProbabilitySimple RandomSystematicStratifiedClusterMulti-StageConvenienceJudgmentSnowball/ReferralQuotaSimple Random and Systematic samplingSimple random sampling is a sampling method in which each element of the population has an equal probability of being selected.Systematic sampling is a sampling process that involves randomly selecting an initial starting point on a list, and thereafter every nth element in the sampling frame.Stratified SamplingStratified sampling requires the researcher to partition the target population into relatively homogeneous subgroups that are distinct and non-overlapping.Proportionate: the number of elements chosen from each of the strata is proportionate to the size of a particular strata relative to the overall sample size.Disproportionate: the number of elements chosen from each of the strata is not based on the size of the stratum relative to the target population size, but rather is based either on the importance of a particular stratum or its variability.Cluster Sampling and Multi-stageCluster sampling is a form of probability sampling in which the relatively homogeneous individual clusters where sampling occurs are chosen randomly and not all clusters are sampled.Cluster sampling involves dividing the population into clusters and randomly selecting a pre-specified number of clusters and then either collecting information from all the elements in each cluster or a random sample. With multi-stage cluster sampling the same process is completed two or more times.Non-probability Sampling MethodsConvenience sampling involves selecting sample elements that are most readily available to participate in the study and who can provide the required information.Judgmental sampling is a form of convenience sampling, sometimes referred to as a purposive sample, in which the researcher’s judgment is used to select the sample elements.Quota sampling is similar to proportionately stratified random sampling but the selection of the elements from the strata is done on a convenience basis.Snowball also called a referral sample, the initial respondents typically are chosen using probability methods and these respondents then identify others in the target population.Factors to consider in determining sample sizeThe variability of elements in the target populationTime available Budget Required estimation precision Whether findings will be generalizedStatistical Factors to considerThe degree of confidence (often 95%)The specified level of precision (amount of acceptable error)The amount of variability (population homogeneity)Calculation of minimum sample size Calculation of minimum sample size Cont’d Negotiating Access in ResearchThere are two level of access you need to negotiate:Physical access: Can be difficult because:The organisation may not be willing to engage in additional activitiesThe request may fail to meet the interest of the organisationCognitive accessStrategies to gain AccessAllow yourself sufficient timeUsing existing and new contactsProvide a clear account of purpose and type of access requiredOvercome organisational concernsHighlight possible benefits to the organisationUse suitable languageFacilitate repliesDevelop access incrementallyEstablish credibility