Data Collection Design Seminar 4

Data Collection Design

• Seminar 4: Presented by Dr. Waqas Ahmed
• Agenda Topics: Sampling, Survey, Measurement Scales, Piloting, Ethical Considerations

Sampling

• Definition: Selecting a small group from a larger group to understand the larger group.
• Purpose: Researchers often can't study the entire population, so they select a representative sample.
• Types of Non-Probability Sampling:
  • Convenience Sampling: Selecting participants based on convenience. Assumes the sample mirrors the overall population.
  • Judgment Sampling: Selecting participants based on the researcher's or expert's judgment that they meet specific criteria. Assumes experts' opinions are representative.
  • Quota Sampling: Dividing the population into subgroups (e.g., gender, age). Taking a sample from each to be representative of the subgroups' proportions in the population.
  • Snowball Sampling: Initial participants refer new participants meeting the criteria. Usually used in hard-to-reach populations.
• Types of Probability Sampling:
  • Simple Random Sampling: Each member of the population has an equal chance of being selected.
  • Systematic Random Sampling: Selecting every nth member from a list.
  • Stratified Random Sampling: Dividing the population into subgroups and randomly selecting from each.
  • Cluster Sampling: Dividing the population into clusters and randomly selecting entire clusters.

Survey Research

• Purpose: Collect data from a large sample to describe and/or understand relationships.
• Characteristics: Structured responses, administered to a wider group of people.
• Applications: Make predictions and understand relationships.

Survey Methods

• Types:
  • Personal: Door-to-door, Mall Intercept
  • Mail: Mail Panel, Mail Survey
  • Telephone: Telephone Interview (Computer Assisted)
  • Internet: Email, Online Survey
• Method Comparison (Table): The table details the various survey methods' contact, sample size, depth, length, response speed, cost, and bias.

Measurement Levels

• Nominal: Categorical data. (e.g., gender, marital status)
• Ordinal: Ranked data; Magnitude relationships exist. (e.g., education level, satisfaction level)
• Interval: Numerical data with equal intervals between values. (e.g., temperature)
• Ratio: Numerical data with a true zero point. (e.g., height, weight)
  • The table details the permissible statistics for each measurement level.

Responses

• Nominal: Categorical data, limited response options. (e.g. "do you have trouble training your employees?")
• Ordinal: Ranked data, visual analogue scales (e.g., "How satisfied are you with the product?")
• Likert Scale: Rate agreement with statements using a scale. (e.g., strongly agree to strongly disagree)

Questionnaire Design

• Wording: Use plain language, simple questions, and mostly closed-ended questions.
• Layout: Use a professional look, enough spacing, clear instructions, avoid splitting questions across pages. Consistent and clear instructions. 
• Pilot Testing: Discuss the questionnaire with your supervisor and field-test it with the target group.

Ethical Considerations

• Participant rights: Protecting participant privacy, minimizing burden, maintaining sensitivity to different cultures.
• Research Conduct: Observing professional standards and conducting scientifically rigorous data collection and analysis.
• Data Report and Access: Appropriately reporting findings and ensuring proper access to data.
• Quality control: Establishing and adhering to research procedures. Documenting all relevant ethical procedures.

Validity and Reliability

• Validity: Measures what it claims to measure.
• Reliability: Results are consistent over time and across respondents.