Lec 9 - Collecting Data: Methods, Tools, and Issues

Data Collection

• Process of gathering data in order to answer a research question

Primary Data

• Data is provided directly by the sources

• Data is obtained directly from the sources

Secondary Data

• Data are obtained or collected by another person for research purposes

• Data is initially collected for other purposes but may be used for research purposes

Survey

• Composed of a series of questions posed to a group of subjects

• Used for descriptive or subjective data about constructs such as:

  • Perceptions, fears, motivation, attitude, outcomes, characteristics, current practices

• If standardized, can be used to assess outcomes related to:

  • Functions, health, status, and quality of life

• May be used for experimental, exploratory, descriptive, and qualitative

Interview

• Researcher verbally asks respondents specific questions and records their answers for later analysis

• May be:

  • Key information interview, one-on-one

  • Focus group discussion, small group with one facilitator

• Methods:

  • Face-to-face

  • Over the telephone

  • Videoconferencing 

Considerations: rapport-building with respondents

Structured Interview

• Standardized set of questions

• All respondents are exposed to the same set of questions

• Same order of questions, with the same choices

Unstructured Interview

• No fixed agenda

• Proceeds with informally with question

• Discusses issues and concerns

• Conversational

• Often used for qualitative studies

Interview bias

• Change in the responses of the interviewee due to self-interpretation

Questionnaire

• Structured surveys that are self-administered using pen and paper or electronic form

• Distributed via:

  • Mail

  • Personal distribution

  • Electronic distribution

• Useful for studying phenomena that can be assessed through self-observation, such as attitudes and values

• NOT useful in studying behaviors that require direct observation

Self-Report Method

• Done on their own, self-administered

• Based on respondent’s report and not from direct observation

• Generally valid for constructs not readily measured by other means

• Has potential for bias or inaccuracy

  • Recall bias, participants having a modelled memory

  • Participants might answer what they think is correct rather than what they know is true

Observation

• Systematic use of senses to gather information for analysis

• Involves understanding and making sense of what is observed

• Useful for qualitative studies

• Can be recorded for later viewing and analysis

• Can be recorded through:

  • Observation quides

  • Recording sheets or checklists

  • Field notes

  • Pictures

  • Combinations 

Overt Observation

• People know that they are being observed

• Downside: Hawthorne Effect

  • People modify their behavior because of awareness of being watched 

Covert Observation

• People do not know that they are being observed.

• Raises ethical considerations 

Participant Observation

• Researcher takes part in the activity being observed

• Allows appreciation of the viewpoint of those being observed

• Enhances validity as the researcher experiences the situation firsthand

Non-participant Observation

• Researcher does not take part and observes quietly from a distance

• Observer remains uninvolved in the activity

Structured Observation

• Uses a checklist of things to observe

• A checklist of predetermined evidence to be observed in the situation, which will either support or refute a preconceived idea

Unstructured Observation

• No specific checklist or predefined evidence

• Events, issues, and phenomena naturally emerge during observation

Analysis of Existing Data

• Analysis of information that was either gathered by someone else or for some other purpose than the one currently being considered

• Sources of data may be:

  • Journals, books, public records, university or hospital records, transcriptions, data sets, videos

• Analysis is done in the form of:

  • Reviews, audits, or recored review

Data Extraction/Abstraction Form

• When analyzing existing data, you need a tool to help extract the data

• Could be in:

  • Paper form

    • Cost-effective

    • Easier to bring across sites

  • Electronic form

    • Used for large-scale data collection

    • Easy to store, centralize, and analyze

    • Customizable, allows for coded responses

Use of Tests and Measures

• Use of specific tests and measures, especially in clinical practice

• May be experimental

• Often applied to describe a phenomenon or determine a cause-and-effect relationship

• Used to assess functional and performance outcomes

Appropriateness to client group

• Tool should match the participant’s age, comprehension level, and ability to respond

Practicality of administration

• Use of data collection tools that are efficient and practical to administer

• Best form of delivery to easily reach participants

Psychometric properties

• What needs to be measured should be accurate and reliable

  • Measures what it intends to measure = validity

  • Produces consistent results = reliability