srd lecture 1 and 2

• Purpose of research:

  • Information gathering:
  • Exploratory
  • Descriptive
  • Theory testing:
  • Explanatory
  • Predictive
• Main pros and cons of a survey as a standardised stimulus used in social science measurement:

  • Pros:
  • Ecologically valid
  • Cost-efficient
  • Can obtain lots of data
  • Cons:
  • Low compliance
  • Reliance on self-report
• Survey administration methods:

  • Self-administered: postal, delivered, online (email, web, mobile)
  • Interview-administered: phone and face-to-face structured interview
• Pros of self-administered and interview-administered surveys:

  • Lower cost/time
  • Less demand characteristics
  • Access to a representative sample
  • Anonymity
  • Higher response rate
  • High data quality and richness
  • Can accommodate cultural differences
• Cons of self-administered and interview-administered surveys:

  • Non-response and low response data
  • Quality and richness of data
  • Adjustments for children, people with lower literacy
  • Time/cost of data collection and entry
  • Demand characteristics
  • Access to representative or dispersed sample
• Pre-testing a survey:

  • Recruit a small convenience sample
  • Have participants talk through their thoughts on each question and give feedback
  • Revise items that don't apply to everyone, are redundant, misunderstood, or incomplete
  • Reconsider ordering and layout
• Pilot testing a survey:

  • Recruit a small sample from the target population
  • Analyze the data and revise the survey based on feedback
• Participant information in a survey:

  • Includes the study name, purpose, required participation, voluntary nature, risks/costs/rewards, use of results, ethics approval number, and contact details
• How to get consent in a survey:

  • Active consent: participant must indicate whether they consent to participate
  • Passive consent: continuing onto the survey implies consent
• Ethical considerations in a survey:

  • Informed consent
  • Minimize risk of harm
  • Confidentiality or anonymity
  • No coercion
  • Minimal deceit & full debrief
  • Honor promise to provide results
  • Be aware of bias or conflicts of interest
• Types of questions in a survey:

  • Objective: a true answer exists
  • Subjective: asking about personal perceptions
• Open-ended vs closed-ended questions in a survey:

  • Open-ended: useful for gathering rich information but more difficult to analyze
  • Closed-ended: good for hypothesis testing and easy to analyze
• Closed response formats in a survey:

  • Dichotomous, multichotomous, multiple response, ranking, verbal frequency, Likert, semantic differential, graphical, non-verbal
• Things to avoid in a survey:

  • Double-barreled questions
  • Double-negative questions
  • Leading questions
  • Loaded questions
• Avoid double-barreled questions that contain more than one concept or purpose, as they can confuse respondents
• Avoid double-negative questions that create confusion in responses
• Avoid leading questions that suggest the answer the researcher is looking for
• Avoid loaded questions that suggest socially desirable answers or are emotionally charged
• Open-ended questions in a survey provide space for answers, useful for gathering rich information but are more difficult to analyze
• Closed response formats in surveys include:

  • Dichotomous: choose from two response options.
  • Multichotomous: choose from multiple response options.
  • Multiple response: choose all that apply.
  • Ranking: measures the relative importance of several items.
  • Verbal frequency: rates the frequency of an event.
  • Likert scale: measures strength of feeling or perception.
  • Semantic differential: puts two words at opposite ends with interval marks.
  • Graphical: marks response with a cross on a line.
  • Non-verbal: participants point to a face that shows how they feel
• Levels of measurement:

  • Nominal: attributes are named, no meaningful order.
  • Ordinal: attributes can be ordered.
  • Interval: distance is meaningful, zero is arbitrary.
  • Ratio: conveys order and distance, with a meaningful zero
• Sampling terms:

  • Target population: who to generalize to.
  • Sampling frame: who can be selected.
  • Sample: who was selected and responded.
  • Representativeness: extent sample is a good indicator of target population
• Sampling methods:

  • Probability sampling: each member has an equal chance of selection.
  • Non-probability sampling: useful for exploratory research or case studies.
  • Simple random, systematic random, stratified random: types of probability sampling.
  • Convenience, purposive, snowball: types of non-probability sampling
• Fisher’s exact test is used to determine if there is a significant association between two categorical variables, like in the Lady tasting tea test
• Student’s t-test is used to determine if there is a significant difference between the means of two independent groups (independent samples t-test) or related groups (paired samples t-test)
• To use the Student’s t-test, the dependent variable must have interval or ratio data, where the distances between each value on the scale are equal and meaningful
• Frequentist statistics focus on observed frequencies and treat probabilities as equivalent to the number of times an event occurs, using p-values and confidence intervals
• Bayesian statistics, on the other hand, is used to update beliefs about the probability of events based on new/existing information, often in problems where prior knowledge is available
• Positivism is a philosophical approach that emphasizes the scientific method and the use of empirical evidence as the basis for knowledge and understanding of the world
• The scientific method, closely tied to positivism, involves observing phenomena, asking questions, making hypotheses, conducting experiments, collecting data, analyzing data, interpreting and concluding, and applying findings
• Three main types of research methods are experimental, quasi-experimental, and non-experimental, each with its own characteristics and purposes
• Survey research is a standardized stimulus and measuring instrument used widely in social sciences, providing quantitative or qualitative data and aiming for impartial sampling and ecological validity
• Survey research can be used for information gathering or theory testing and building, with purposes ranging from exploratory to predictive research
• Constructs measured in surveys can include personality traits, perfectionism, procrastination, planned study time, and expected marks
• When writing a research question, it's important to consider the cross-sectional nature of the data and ensure it can be answered sensibly with the survey data
• After formulating a research question, writing a hypothesis based on previous research findings is crucial, showing an expectation of the answer and citing relevant sources
• Advantages of survey research include ecological validity, access to a wide range of participants, potentially large amounts of data at a relatively low cost, and ethical considerations, while disadvantages include lack of control, less internal validity, superficial data, and potential high costs for representative data