Please enable JavaScript.

Coggle requires JavaScript to display documents.

COLLECTING DATA (Quantitative data (What permissions will you need?…

- - - - level
        
        one for dependent variable
        
        one for independent variable
        
        e.g. individual, family, school,
        
        level at which data will be gathered
    - - population
        
        group of individuals that comprise the same characteristics
      - sample
        
        sub-group of target population
        for our study
        to help make generalizations about the population
        
        are only estimates
        
        sampling error = true population - sample estimate
      - sampling
        
        probability sampling
        
        select individuals such that they represent the population
        
        types
        
        SIMPLE RANDOM sampling
        
        all in the population has an equal chance of being selected in the sample
        
        STRATIFIED sampling
        
        stratify population on a characteristics (e.g. gender), then sample each stratum
        
        MULTI-STAGE CLUSTER sampling
        
        sample chose in 1 or 2 stages because population not easily identified
        
        SYSTEMATIC sampling
        
        choosing every "nth", individual or site in the population until the desired sample size is achieved
        
        non-probability sampling
        
        select participants that are available, convenient or have some characterisitcs that researcher are interested in
        
        types
        
        CONVENIENCE sampling
        
        participants are available & willing to be studied
        
        SNOWBALL sampling
        
        researcher ask participants to help identify other potential participants to become members of the sample
  - - - tool to measure, observe, or document quantitative data
      - types
        
        performance measure e.g. test performance
        
        Attitudinal measure e.g. measure feelings
        
        Behavioural measures e.g. observe behaviour
        
        factual measure e.g. documents, records
      - flow
        
        identify variable
        
        define variable
        
        locate data (measures,
        observations, documents with Q's and scales
        
        collect data on instruments
        yielding numeric scores
  - - - ERIC search - use the term 'instruments' and topic of study
      - ERIC website for evaluation & assessment
      - develop own instruments
      - examine guides to commercially available tests
      - published journal articles
    - - scales of measurement
        
        Ordinal (categorical)
        
        Interval (continuous)
        
        ratio (continuous)
        
        Nominal (categorical)
      - VALIDITY
        
        scores from measuring variables that are meaningful
        
        types
        
        content - representative of all Q's to be asked
        
        criterion referenced
        
        construct
        
        refers to how well a test or tool measures the construct that it was designed to measure e.g. to what extent does the BDI measures depression?
      - RELIABILITY
        
        scores from measuring variables that are stable & consistent
        
        e.g. bathroom scale
        
        types
        
        test-retest i.e. scores are stable over time
        
        alternate forms (equivalence of 2 instruments)
        
        inter-rater reliability
- - - - select people & sites who can best help to understand our phenomenon
        
        might give useful information
        
        that might help people to learn about the phenomenon
        
        might give voice to silence people
      - BEFORE sampling
        
        TYPES
        
        Homogenous sampling
        
        to describe sub group in depth
        
        Typical sampling
        
        maximal variation sampling
        
        theory or concept sampling
        
        extreme case sampling
        
        describe particular troublesome/enlightening cases
      - AFTER sampling
        
        TYPES
        
        opportunistic sampling
        
        to take advantage of whatever case unfolds
        
        snowball sampling
        
        to locate people/site to study
        
        confirming/ disconforming sampling
    - - 1 individual
      - small for in-depth perspective
      - 4 cases
      - 20 - 30 interviews
      - Delphi technique
  - - - process of gathering first-hand info by observing people & places at research site
      - roles
        
        participant observer
        
        non-participant observer
      - observational roles can change
      - do multiple observations
      - record both descriptive & reflective field notes
    - - types
        
        personal
        
        phone
        
        email
        
        focus group
      - general, open ended Q's
      - Info is RECORDED & TRANSCRIBED
      - unstructured
    - - public & private records
      - good source for text data
      - obtain permission before using documents
      - optically scan documents
    - - determine materials that can provide evidence
      - available? need permission
      - check accuracy & authenticity
      - collect & organize data
  - - - Interview protocols
        
        HEADER: info about interviewer
        
        open ended questions
        
        incld. ice breaker
        
        incld. ones that address major research Q's
        
        probes that clarify & elaborate
        
        closing comments & thank participant
      - observation protocols
        
        header: same as interview
        
        left column: records descriptive notes
        
        right column: record reflective notes
        
        sketch picture of the site
  - - - limit initial collection to 1-2 observations/interviews
      - time needed to create a substantial database
    - - anonymity of participants
      - convey true purpose of study (don't conceal) without deception