Welcome
Tutorials PLEASE: Turn off your cell phone Be on time for the class Attend regularly Come prepared
Dates of tutorials
The research process Stage 1: Definition of the problem Stage 2: Obtaining the information Stage 3: Analysing and interpreting the data Stage 4: Communicating the results
The research process STAGE 1: DEFINING THE PROBLEM Deciding on the research topic Conducting a literature review Specify a research question Formulating a hypothesis Operationalizing concepts
The research process STAGE 2: Obtaining the information Ethics Research design Sampling Data collection STAGE 3: Analysing and interpreting the information Describing and interpreting quantitative data Analyses and interpretation of qualitative data
The research process STAGE 4: Communicating the results Report
Sampling
Outcomes Key terms relating to sampling Reasons for sampling Probabilility and nonprobability sampling Techniques in probability sampling Techniques in nonprobability sampling Factors that influence the sample size Drawing a sample in practice
Basic sampling concepts Method (process) of selecting certain members to represent the whole group Obtaining info from a few cases and then drawing conclusions from that to the population 2. Population The entire group of persons (or set of objects / events) of interest to the researcher Also called target population eg single parents
Basic sampling concepts Researcher should describe the population in terms of criteria: Eligibility, inclusion criteria, distinguishing descriptors eg readers must be woman and they must be frequent readers of that specific women’s magazine Another way of clearly defining a population is through an operational definition of the population for eg family violence Accessible population or study population – the population the researcher has access to
Basic sampling concepts 2. Element / unit of analysis Is a case or unit from a defined population about which information is obtained People, objects, events or social groups All elements together constitute the population 3. Parameter Is a characteristic of the elements of a population eg the age of the students being researched Describes a particular characteristic of the whole population
Basic sampling concepts 4. Sample Is a subset of a population 5. Sampling frame A comprehensive list of all units (elements)from which the sample is drawn 6. A representative sample A sample that resembles the population Which enables the researcher to accurately generalize the results
Basic sampling concepts 7. Sampling error Refers to the differences between population parameters (eg the average age of the population) and sample statistics (the average age of the sample)
Basic sampling concepts Sampling error May occur because of: Chance factors – one element may have been included rather than another Bias in selection – faulty technique Non-response errors – when an element of the population does not respond to a measurement instrument
Basic sampling concepts Sampling bias Is when a sample is not representative of the population Is caused by an incorrect selection process Is an over or underrepresentation of a segment of the population which will then impact on validity of study Is a threat to external validity when the subjects are not randomly selected from the population
2 Important factors in sampling To draw a representative sample, the following is nb Similarity of population: how similar or dissimilar is the population? The degree of precision with which the population is specified
2 Important factors in sampling To draw a representative sample, the following is nb Similarity of population Heterogeneous population - consists of people who are dissimilar to each other Homogeneous population – consists of people who are similar to each other eg a group of female junior tennis players who play in the same league and for the same province The more alike the elements of the population, the smaller the sample can be and still be representative
2 Important factors in sampling To draw a representative sample, the following is nb 2. Defining of population The degree of precision with which the population is specified The sampling frame is the defined population from which the sample is drawn Eg all indv currently on ARV treatment for HIV
Sampling approaches Probability or random sampling Nonprobability sampling
Probability / random sampling Each person has an equal chance of being selected for the sample Also called random sampling – the selection of elements is based on some form of random procedure Forms of probability / random sampling techniques: Simple random sampling Systematic sampling Stratified random sampling Cluster sampling
Simple random sampling Each element has an equal chance of being in the study To prevent sample bias Steps in simple random sampling First the population is defined Then the sample frame is drawn up Each element of the sample frame then has an equal change of being included in the sample 4. Random selection techniques Lottery or fishbowl techniques A random table number – use a table of random numbers to select the subjects
Systematic sampling Also called interval sampling It is when a sampling frame is available Involves drawing every fth element from a population Elements are selected at equal intervals eg every 6th, 10th or 19th element
Systematic sampling Procedure: Obtain a list of the total population – N: is the symbol for the size of the population. Determine the sample size (n) Determine the sampling interval (k) by dividing the size of the population by the size of the sample. Choose a random starting point between 1 and k. p158
Stratified random sampling The population is divided into different groups or subgroups called strata, so that each element of the population belongs to one and only one stratum Random sampling is then done within each stratum, using either simple random sampling or systematic sampling.
Stratified random sampling Two ways of determining the number of elements sampled from each stratum Proportional stratified samples The number of elements selected from each stratum is proportional to the size of the stratum in the population 3. Disproportional stratified sampling The number of elements selected from each stratum is not proportional to the size of the stratum in the population
Cluster sampling Requires that the population is divided into groups or clusters Grouped in heterogeneous clusters Used when a sampling frame is not available Takes place in successive stages – multistage sampling Researcher samples a population that is much more general than the final one (cities) Then progresses to the next most inclusive sampling units (residential blocks) Then reaches the final stage (participants in study)
Cluster sampling Tend to contain more sampling errors than simple or stratified random sampling Is considered more economical and practical than other types of probability sampling
Nonprobability sampling Where we don’t know whether we have included each element of the population in a sample Cannot generalize our results to the general population Used when probability sampling is extremely expensive or difficult or representativeness is not essential, when no sampling frame is available Where the aim is to generate theory and to gain a wider understanding of social processes
Nonprobability sampling Convenience or accidental sampling Quota sampling Purposive or jugdmental sampling Snowball sampling
Nonprobability sampling Convenience or accidental sampling When the researcher selects those elements that he or she can access easily until the sample reaches the desired size Also called accidental or availability sampling Eg people in library available at a certain time Quota sampling Similar to stratified sampling, except that the final selection of elements is not random Sampling procedure relies on convenience or accidental choice
Nonprobability sampling 3. Purposive or judgmental sampling Also known as theoretical sampling The researcher selects a sample that can be judged to be representative of the total population Commonly used in qualitative research 4. Snowball sampling Research respondents obtaining other potential respondents
Sample size Refers to the number of elements in a sample The larger the sample, the more valid and accurate the study The more heterogeneous a population is, the larger the sample must be to properly represent the characteristics of the population See factors that influence sample size on p165