Data Collection Methods and Classification

• Data collection is covered in this unit.

Expected Learning Outcomes

• Outline the steps involved in a scientific inquiry.
• Classify different types of data.
• Identify different sources of data.
• Differentiate between census and survey.
• Differentiate the different methods of data collection like survey, observation, and experimental methods.
• Distinguish between probability and non-probability sampling and procedures.

Steps in Scientific Inquiry

• Define the problem.
• Formulate the research design.
• Collect the data.
• Code and analyze the collected data.
• Interpret the results.

Classification of Statistical Data

Primary vs. Secondary Data

• Primary data is measured by the researcher/agency that publishes it.
• Secondary data is a republication of data by another agency.
• Publications from the Philippine Statistics Authority is primary data.
• Subsequent publications of the same data by other agencies is secondary data.

External vs. Internal Data

• Internal data relates to the operations and functions of the organization collecting the data.
• External data relates to some activity outside the organization collecting the data.
• Sales data of SM is internal data for SM but external data for another organization like Robinson's.

Sources of Statistical Data

General Classification of Collecting Data

• Census (complete enumeration) gathers information from every unit in the population.
• It is not always possible to get timely, accurate, and economical data.
• It is costly when the number of units in the population is too large.
• Sample survey gathers data from a small but representative cross-section of the population.

Advantages of Sample Survey over a Census (on a Large Population)

• Speed and timeliness: data can be gathered faster, ensuring uniformity.
• Economy: information gathering and data analysis is cheaper.
• Quality and accuracy: a sample survey yields more accurate results because a small, skilled group makes fewer errors.
• Feasibility: some data gathering methods require the destruction of a unit.

Methods of Collecting Statistical Data

Survey Method

• Questions are asked to obtain information.
• Administering a survey includes telephone interviews, mailed questionnaires, and online surveys.
• Telephone interviews encourage response, allow interviewers to clarify questions, are cost-efficient for national sampling frames, and could have interviewer bias.
• Telephone interviews cannot be used for non-audio information and are unreliable in rural areas with low telephone penetration.
• Types of telephone interviews: traditional, computer-assisted telephone dialing (CATI), and computer-assisted telephone interviewing.
• Mailed questionnaires may be handed to or mailed to respondents, returned via mail.
• Mailed questionnaires are low-cost.
• Mailed questionnaires have long delays before surveys are returned so are not suitable for issues needing clarification.
• Mailed questionnaires allow respondents to answer at their own convenience and introduce no interviewer bias.
• Online surveys can use web or e-mail, with web preferred for using interactive HTML forms.
• Online surveys are inexpensive, fast, easy to modify, but honesty of responses can be an issue.
• Online surveys are easy to manipulate and can be automated for data creation and reporting.
• Online Surveys are used in large-scale industries and create data sets in real time, but are sometimes incentive-based.
• Determining/controlling selection probabilities is often difficult, hindering inferential analysis of data.
• Personal in-home surveys involve interviewing respondents in person in their homes.
• Personal in-home surveys are very high cost.
• Personal in-home surveys are suitable for locations where telephone or mail is not developed.
• Skilled interviewers in personal in-home surveys can improve response rates, but there is potential for interviewer bias.
• Personal mall intercept surveys intercept shoppers and interview them on the spot, or give them a questionnaire to complete.
• People feel a mall is a more appropriate place to do research, but interviewer bias is possible.
• Mall intercept surveys can be manipulated by completing multiple times to skew results.
• Observation method allows the recording of behavior only at the time of occurrence.
• Observation does not rely on the respondent's willingness to provide information.
• Data like behavior patterns can be collected only by observation
• The potential bias caused by the interviewing process is reduced or eliminated.
• Observation disadvantages: things such as awareness, beliefs, feelings, and preferences cannot be observed.
• Observed behavior patterns can be rare or too unpredictable, increasing data collection costs and time requirements.

Experimental Method

• Designed for collecting data under controlled conditions.
• An experiment is an operation where there is human interference with the conditions that can affect the variable under study.
• This is an excellent method of collecting data for causation studies.
• Properly designed experiments will reveal the effect of a change in one variable on another with accuracy.

Other Methods

• Utilizing existing studies like census, health statistics, weather bureau reports, etc.
• Documentary sources: published or written reports, periodicals, unpublished documents, etc.
• Field sources: researchers who have done related studies are asked personally or directly for information.
• Registration method like car registration, student registration, and hospital admission.

Sampling and Sampling Techniques

• Probability sampling gives every element of the population a known, nonzero chance of being selected in the sample.
• Non-probability sampling does not give all members of the population a chance to participate in the study.
• The selection of samples is based on the subjective judgment of the researcher rather than random selection.
• Probability sampling is used because there is no objective way of assessing the reliability under non-probability sampling.
• Target population: the population from which information is desired.
• Sampled population: the collection of elements from which the sample is taken.
• Population frame: a listing of all the individual units in the population.

Non-probability Sampling Methods

• Often used in exploratory and qualitative research to develop an initial understanding.
• Purposive sampling: sets out to make a sample agree with the profile of the population.
• Quota sampling: selects a specified number of sampling units possessing certain characteristics.
• Convenience sampling: selects sampling units that come to hand or are convenient to get information from.
• Judgment sampling: selects sample in accordance with an expert's judgment.
• Snowball sampling: existing subjects provide referrals to recruit samples required for a research study.

Probability Sampling Methods

• Often used in quantitative research. If the aim of the research is to produce representative results.
• Simple random sampling.
• Stratified random sampling.
• Systematic sampling.
• Cluster sampling.
• Multistage sampling.

Simple Random Sampling

• Method of selecting n units out of the N units where every distinct sample of size n has an equal chance of being drawn.
• The process of selecting the sample must give an equal chance of selection to any one of the remaining elements in the population at any one of the n draws.
• Random sampling may be with replacement (SRSWR) or without replacement (SRSWOR).
• In SRSWR, a chosen element is always replaced before the next selection.
• n is distinct for SRSWOR, not necessarily distinct for SRSWR.

Steps for Simple Random Sampling

• Step 1: Make a list of the sampling units and number them from 1 to N.
• Step 2: Select n numbers from 1 to N using some random process.
• Step 3: The sample consists of the units corresponding to the selected random numbers.

Advantages of Simple Random Sampling

• Easier theory.
• Simple inferential methods.

Disadvantages of Simple Random Sampling

• The sample chosen may be widely spread, entailing high transportation costs.
• A population frame is needed.
• Results in less precise estimates if the population is heterogeneous.

Stratified Random Sampling

• The population of N units is first divided into subpopulations called strata.
• Then a simple random sample is drawn from each stratum, the selection being made independently in different strata.

Steps for Stratified Random Sampling

• Step 1: Divide the population into strata.
• Ideally, each stratum must consist of more or less homogeneous units.
• Step 2: After the population has been stratified, a simple random sample is selected from each stratum.

Proportional Allocation

• A decision is required on how to divide a fixed sample size among the different strata.
• Select a number of units from each stratum that is proportional to the number of sample units within each stratum.
• A proportional allocation is a common strategy.
• The formula in calculating the sample size (n sub i) for each stratum under proportional allocation is given by: n sub i = n (N sub i/N).
• n is sample size from population, N is population size, n sub i is the stratum sample size, and N sub i is the stratum size.
• Other allocation techniques include equal and optimal allocation.

Advantages of Stratified Random Sampling

• Stratification may produce a gain in precision in the estimates of characteristics of the population.
• It allows for more comprehensive data analysis since information is provided for each stratum.
• It is administratively convenient.

Disadvantages of Stratified Random Sampling

• A listing of the population for each stratum is needed.
• The stratification of the population may require additional prior information about the population and its strata.

(1-in-k) Systematic Sampling

• Selecting a sample by taking every kth unit from an ordered population.
• The first unit is selected at random.
• k is the sampling interval; 1/k is the sampling fraction.

Steps for Systematic Sampling (Method A)

• Step 1: Number the units of the population consecutively from 1 to N.
• Step 2: Determine k, using the formula k = N/n.
• Step 3: Select the random start r, where 1 is less than or equal to r which is less than or equal to k.
• The unit corresponding to r is the first unit of the sample.
• Step 4: The other units of the sample correspond to r + k, r + 2k, r + 3k, and so on.

Steps for Systematic Sampling (Method B)

• Step 1: Number the units of the population consecutively from 1 to N.
• Step 2: Let k be the nearest integer less than N/n.
• Step 3: Select the random start r, where 1 is less than or equal to r which is less than or equal to N.
• The unit corresponding to r is the first unit of the sample.
• Step 4: Consider the list of units of the population as a circular list.
• The other units in the sample are the units corresponding to r + k, r + 2k, r + 3k,..., r + (n-1)k.

Advantages of Systematic Sampling

• Easier draw and execute without mistakes than simple random sampling.
• Possible to select a sample in the field without a sampling frame.
• The systematic sample is spread evenly over the population.

Disadvantages of Systematic Sampling

• If periodic regularities are found in the list, a systematic sample may consist only of similar types.
• Knowledge of the structure of the population is necessary for its most effective use.

Cluster Sampling

• Selecting a sample of distinct groups, or clusters, of elements and then taking a census of every element in the selected clusters.
• Similar to strata in stratified sampling, clusters are non-overlapping subpopulations, which together comprise the entire population.
• Clusters are preferably formed with heterogeneous, rather than homogeneous elements.
• Clusters may be of equal or unequal size.
• When all of the clusters are the same size, the number of elements in a cluster will by denoted by M, while the number of clusters, N

Steps for Cluster Sampling

• Number the clusters from 1 to N.
• Select n numbers from 1 to N at random.
• The clusters corresponding to the selected numbers form the sample of clusters.
• Observe all the elements in the sample of clusters.

Advantages of Cluster Sampling

• No population list of elements needed.
• Reduced listing cost.
• Reduced transportation cost.

Disadvantages of Cluster Sampling

• The cost and problems of statistical analysis are greater.
• Estimation procedures are more difficult.

Multistage sampling

• Multistage sampling divides the population into a hierarchy of sampling units corresponding to the different sampling stages.
• First stage: Population is divided into primary stage units (PSU), then a sample of PSUs is drawn.
• Second stage: Each selected PSU is subdivided into second-stage units (SSU), then a sample of SSUs is drawn.
• Carried to a 3rd, 4th stage, and so on, by sampling the subunits instead of enumerating them completely at each stage.

Advantages of Multistage Sampling

• Reduced listing cost and reduced transportation costs.

Disadvantages of Multistage Sampling

• Estimation procedure is difficult, especially when the primary stage units are not of the same size.
• Estimation procedure gets more complicated as the number of sampling stages increases.
• Sampling procedure entails much planning before selection is done.
• Multistage sampling may use any probability sampling at any sampling stage.

Explore data collection methods, sources, and types. Learn the steps in scientific inquiry. Primary data is directly measured, while secondary data is republished. Differentiate between external and internal data.