Sampling Theory free essay sample

Questioning error, Recording Error, Interference Error Data Error, Which could be intentional, Unintentional Failure to contact all, Incomplete responses  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 7 ? It is critical to determine the target population ? To eliminate the Specification error as well as sampling frame error ? A cliche, however, definition of Research objective is critical ? Optimal definition of population ? Most research failures suffer from this problem! ? This clarity (or lack of it) impacts the questionnaire design ? Too fine and it is too restrictive, expensive and operationally difficult ? Too broad and might confound the findings ? But convenience not at the expense of appropriate research design ? Convenience is critical  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 8 ? Sampling frame ? Selection of the list ? Telephone Directory ? In most cases, it works fine as it provides a complete list. ? Sometimes it may not be complete (non-selection error) ? In the US, Presidential candidate was predicted on the basis telephone interviews ? MR firms (systematic sampling) Select a neighbourhood/ blocks in a city ? Every nth house is selected ? For Non-response, select kth house after this one, then repeat the process  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 9 ? Sampling Techniques ? Probabilistic sampling ? Non-probabilistic sampling ? Simple Random Sampling ? Sample members are chosen at random from the population – each member having equal probability of being selected ? Often names are put in a box and selected in random ? Vietnam war and December tags ? Probabilistic Sampling ? Stratified Sampling ? There would be sub-groups in a population ? Ensuring representation from each of these sub-groups  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 10 ? Probabilistic Sampling ? Proportional Stratified Sampling ? Samples for each ‘strata’ is selected proportional to the population in each sub-group ? Proportional ‘representation’ in sample ? Inverse proportional stratified sampling ? Also used in some cases, where a particular group’s views (although smaller) is more important ? Rare, but used in specialised research projects ? E. We will write a custom essay sample on Sampling Theory or any similar topic specifically for you Do Not WasteYour Time HIRE WRITER Only 13.90 / page g. Many ‘area’ studies, it is carried out ? Disproportionate Stratified sampling In some special cases, proportional would mean some groups would be ‘under-represented’  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 11 ? Probabilistic Sampling ? Cluster sampling ? Different from Stratified sampling ? Clusters (groups) are selected at random ? Then all members are selected in those groups ? Faster, cost effective coverage ? Systematic Sampling ? Most MR firms engaging in consumer products research engage in some form ? Target population is identified ? Nth members are chosen (driven by sample requirements and applicability) Krishanu Rakshit, IIM Calcutta 28 September, 2010 12 ? Non-probabilistic Sampling ? ? ? ? Judgmental Sampling (expert sampling) Snowball sampling Convenience Sampling Quota Sampling ? (similar to stratified sampling) ? Multi-stage design ? Different techniques maybe used in conjunction ? Depending on the nature of the problem ? It may evolve during the research programme  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 13 ? Determining Sample Size ? What is an appropriate sample size ? Rule of thumb ? 100 for each subgroup ? Budget constrained ? Comparable (benchmark) studies* ? Population Parameters ? Interval (Confidence) – accuracy of prediction ? Variance in population z.? x X? n z 2? 2 n? 2 SE 28 September, 2010 14  © Krishanu Rakshit, IIM Calcutta ? In case of Stratified Sampling ? Where each strata / group has different (Population) variance ? Where cost of each interview/ questionnaire is different for each strata ? Neyman’s rule suggests: size of ith sample: ? i .? ni ? i ? i k ci ? .? ( i ci .n i )  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 15  © Krishanu Rakshit, IIM Calcutta 28 September, 2010 16