Social Science Research

• Survey research


  A method utilizing standardized questionnaires or interviews to collect systematic data about people's preferences, thoughts, and behaviors
• Survey research formalized by sociologist Paul Lazarsfeld to study the impact of radio on political opinions in the US


  1930s-40s
• Purpose and applicability of survey research

  • Used for descriptive, exploratory, or explanatory research, particularly suitable for individual-level analysis
  • Other units studied include groups or organizations, often using a "key informant" or "proxy," susceptible to respondent bias
• Strengths of survey research

  • Measures various unobservable data: preferences, traits, attitudes, beliefs, behaviors, or factual information
  • Allows remote data collection for large populations, offering convenience and unobtrusiveness
  • Economical in terms of time, effort, and cost compared to other methods like experiments
• Disadvantages of survey research

  • Susceptible to biases such as non-response, sampling, social desirability, and recall bias
• Types of survey research

  • Questionnaire surveys
  • Interview surveys
• Questionnaire surveys


  • Types: Mail-in, group-administered, or online surveys
  • Advantages: Unobtrusive, cost-effective, convenient for respondents
  • Disadvantages: Low response rates, potential for sampling bias, limited for detailed responses
• Question construction in questionnaire surveys

  • Response Formats: Dichotomous, nominal, ordinal, interval-level, continuous
  • Considerations: Clarity, avoidance of negative wording, ambiguity, bias, double-barreled questions, generality, presumption, imaginary scenarios, respondent knowledge
• Question sequencing in questionnaire surveys

  1. Logical flow from least to most sensitive, factual to attitudinal, and general to specific

  2. Transition statements to guide respondents through sections
• Golden rules for questionnaire surveys
• Interview surveys


  • Interviews are personalized data collection methods conducted by trained interviewers using standardized questions
  • Interview scripts may contain additional instructions for interviewers and space for personal observations
• Types of interview surveys

  • Personal interviews: Face-to-face interactions, conducted at home or office, enhancing response rates but requiring skilled interviewers
  • Group interviews (focus groups): Facilitated discussions among small groups, useful for exploring complex issues but prone to dominance and reluctance
  • Telephone interviews: Utilizing random or computer-assisted dialing techniques, offering higher response rates but limited to audio communication
• Role of interviewer


  • Preparation: Training in methodology, understanding study purpose, and familiarity with potential biases
  • Engagement: Locating respondents, motivating cooperation, and clarifying concerns
  • Facilitation: Guiding interviews, probing for detailed responses, and observing non-verbal cues
• Conducting interviews


  1. Preparing a kit with necessary documents, introducing the study confidently, assuring confidentiality, and obtaining permission for recording

  2. Following questionnaire scripts precisely, avoiding changes in wording or question sequence, and using probing techniques for detailed responses

  3. Post-interview actions include thanking respondents, providing result expectations, and noting key observations
• Biases in survey research

  • Non-response bias: Addressed through advance notification, relevance of content, shorter surveys, endorsements, and incentives
  • Sampling bias: Occurs due to exclusion of certain population groups, affecting generalizability
  • Social desirability bias: Tendency to provide socially acceptable responses, challenging validity
  • Recall bias: Memory-related inaccuracies in responses, mitigated by anchoring memories to specific events
  • Common method bias: Covariance between variables measured simultaneously, addressed through statistical tests or different measurement methods
• In the first two modules of this course, you learned about the importance of theory as a basis for developing a research question, and you were introduced to some key ideas to keep in mind when designing a research study to answer that question. In this, the third module, you will begin to consider the actual study designs that can help to answer a research question.
• In this lesson, we will begin our discussion of specific research study designs by looking at survey research. In Lesson 9, we will continue our discussion with an examination of experimental research. The final lesson of this module will examine case study research and will conclude with an examination of interpretive research.
• Before you start the online lesson material, please read through Chapter 9 in your text. The online lesson will build on the material from that chapter, so you should be familiar with that before you begin to work through the online lesson.
• Lauritsen (1993) analyzed the National Youth Survey to examine the relationship of family and sibling characteristics with juvenile delinquency. Results indicate a positive relationship between types of sibling offending.
• Baker, Wagner, Singer, and Bundorf (2003) conducted a survey to measure the extent of Internet use for health care. Results indicate that many people indicate that they use the internet for health information, but few report that the information they found on the internet influenced their health decisions, and very few report impacts on their utilization of health care.
• Ward and Wackman (1972) conducted a survey study to examine children's attempts to influence their parent's purchasing decisions and parental yielding to that influence. Results show that children's attempts to influence their parents decrease with age, but that their mother's yielding to the requests increases with the child's age.
• Survey research involves the use of standardized questionnaires to collect data from people about their attitudes, preferences, and behaviours. Questions on a survey may be structured or unstructured but, typically, all respondents in a survey study answer the same questions. Survey research may be conducted using pencil and paper measures, online questionnaires, or in an interview format.
• In this study, Dr. Confidante administered an online questionnaire to a broad sample of university students in Ontario. He carefully developed the survey instrument, and all questions were structured, so that participants selected their responses from a predetermined list. He traveled to five university campuses to invite students to participate in his study in person.
• Because the goal of the research is to evaluate a very large sample, to broadly measure attitudes and behaviours as students are making friends, survey research is the most efficient method for data collection.
• Using an online survey for this study allows participants to complete the survey in a location where they are comfortable, and at a time that suits them.
• Using an online survey helps to reduce errors in data entry which might occur if paper copy surveys are used.
• This technique results in a large sample, which allows for the detection of small effects using statistical analyses.
• The survey used in this study asks students to recall events from the past. For many, their strongest friendships could have been formed several years prior to the research. This may result in recall bias, as some students would be unable to accurately recall their attitudes and behaviours at that time.
• Samuel is interested in evaluating the response of high school students to the lunch program implemented in their school. A group-administered survey would be the best fit.
• Andrea is studying the retirement plans of Canadians. She would like to contact individuals from across the country, to ask about their plans. A self-administered mail survey would be the best fit.
• Mei is evaluating university student response to new dorm rules and guidelines. She believes that this is a complex topic, and would like to hear student experiences, and their responses to the experiences of others. A focus group would be the best fit.
• Guy is working with Family and Children's Services to assess a new parenting program. He is concerned that some of the parents may need help to understand some of the questions in his survey. An interview survey would be the best fit.
• BRUSO is a good way to summarize considerations for survey question construction: Brief, Relevant, Unambiguous, Specific, Objective.
• Respondents must interpret the question, retrieve relevant information from memory, form a tentative judgment, convert the tentative judgment into one of the response options provided, and finally edit their response as necessary.
• Question order can impact respondent answers. When the life satisfaction item comes first, results suggest that the two variables are only weakly related. When the dating frequency item comes first, results suggest that those who date more have a strong tendency to be more satisfied with life.
• Question Order


  One item can change how participants interpret a later item or change the information that they retrieve to respond to a later item
• Response Options


  The response options can also have unintended effects on people's responses
• Order of presentation of candidates' names in an election impacts selection of the candidates. Candidates presented first on the ballot are advantaged by 2.5% (among undecided voters) compared to those presented later on the list.
• Counterbalancing


  The practice of rotating the completion order of equivalent items on a research survey to reduce response order effects
• Items at the beginning of the survey may influence whether participants are likely to recall information in later items, which could affect their responses.