Docstoc

QUANTITATIVE RESEARCH

Document Sample
QUANTITATIVE RESEARCH Powered By Docstoc
					QUANTITATIVE RESEARCH
CONTENT-ANALYSIS
CONTENT ANALYSIS
   I. OVERVIEW
      A. A quantitative method for analyzing the nature,
       functions, & effects of texts.
        1. Originated in 18th century Sweden when a
         group of scholars & clergy analyzed a collection
         of 90 non-orthodox hymns.
        2. Counted religious symbols in hymns to see if
         they were heretical.
        3. Modern biblical exegesis of texts still uses
         content analysis as a method (along with other
         critical methodologies).
CONTENT ANALYSIS
   4. Also used in mass media research.
    a. 1st applied to content of leading
     newspapers.
    b. In 1930s, applied to radio & public
     speeches.
    c. In WWII, invaluable in examining
     German radio transmissions.
    d. More recently, used to examine
     television programs.
 B. Sometimes seen as descriptive, critical work
CONTENT ANALYSIS
   C. Defining content-analysis is confusing & in dispute.
       a. Some define broadly as any technique that
        objectively analyzes messages in a “systematic,
        rule-governed & rigorous way” (Communication
        Research, p. 171).
       b. This definition permits more qualitative
        content-analysis.
       c. Others, however, would not include purely
        qualitative studies as “true” content analysis.
CONTENT ANALYSIS
   3. Frey et. al. on “textual analysis”: a continuum of
    quantitative & qualitative approaches.
     a. More quantitative--content analysis, interaction
      analysis & some historical analysis.
     b. More qualitative--critical analysis (rhetorical or
      media) & other historical analysis.
   4. Standard definition of content analysis:
       Method of studying & analyzing messages in a
        systematic, objective, & quantitative manner for
        the purpose of measuring variables.
CONTENT ANALYSIS
   5. “True” content analysis aims for objective rigor like
    any social science.
     a. Operationalization of concepts.

     b. Classification criteria must be both explicit &
      comprehensive.
     c. Coders essential for more objectivity &
      reliability.
     d. Study should be able to be replicated by
      another researcher with similar results.
CONTENT ANALYSIS
      e. Requires random sampling
     f. Requires equal treatment of all content (through a uniform
       coding process).
   6. Main goal--the accurate & precise description of messages.
     a. Usually done on many texts, not just one.

     b. Quantification aids in making descriptions.

     c. Quantification aids in comparisons among categories &
       sets of data.
     d. Seek generalizable results from sample to population
       (called a universe).
CONTENT ANALYSIS
   D. Functions & uses of content analysis
     1. Description of message characteristics--what,
      how, & to whom something is stated or occurs in a
      text.
       a. To compare different data sets.
       b. To compare content to the “real world.”
       c. To assess images of particular groups in
        society.
       d. To describe cause & effect.
CONTENT ANALYSIS
   2. Explanation of message characteristics:
     a. Examining a dependent variable.

     b. Measuring an independent variable.

     c. Testing hypotheses--relate certain characteristics
      of the source of a given message to its message
      characteristics.
        1) Make inferences about causes--why something
         is stated or occurs in a text.
        2) Make inferences about effects--what happens
         when something is stated or occurs in a text.
CONTENT ANALYSIS
   II. PROCEDURES FOR CONDUCTING A
    CONTENT ANALYSIS
      A. Setting up the study:
       1. Identify the research problem.
       2. Develop RQ(s)
       3. Sometimes develop RH(s)—if testing
         IV/DV relationships will have RH(s).
CONTENT ANALYSIS
       3.   Determine a suitable data base.
            a. Select textual artifacts directly pertinent to
             the research problems.
            b. If can’t do a census of all relevant texts
             must take a representative sample.
   B. Obtaining a representative sample.
     1. Operationalize “population of interest” or
       universe of texts
     2. After defining population boundaries, specify
       the smallest unit or elements to be studied.
CONTENT ANALYSIS
     3.  Construct a sampling frame.
      4. Take a simple, systematic, or stratified random
       sample of texts.
        a. Random sampling--key component.
        b. Can use randomization techniques instead.
   C. Collect data from sample, along with contextual
    information:
      1. Relevant data about extrinsic matters, such as
       source data, historical background, documented
       effects, etc.
CONTENT ANALYSIS
     2.  Contextual information can emerge from
      comprehensive literature review.
     3. Can also emerge through audience surveys, or
      interviews with the source(s), looking at historical
      data, etc.
     4. Helps justify inferences made from the intrinsic
      data analysis.
   D. Develop a measurement system
     1. Unitizing--identifying the units of analysis (or
      message elements).
CONTENT ANALYSIS
  2.  Five common units:
    a. Physical--actual kinds of texts, speeches,
     shows, etc.
    b. Syntactical--symbols, metaphors, words, etc.
    c. Referential --what the text is about.
    d. Propositional--specific positions taken in the
     text.
    e. Thematic--specific topics or themes.
CONTENT ANALYSIS
   3. Count number of times the relevant units appear in
    a text.
   4. Place units into appropriate categories:
     a. Pre-formulated--categories drawn from
      research.
     b. Emergent--categories emerge from the data
      itself.
   5. Define category boundaries with maximum detail
    (helps with reliability).
CONTENT ANALYSIS
   6. Criteria for categorizing:
     a. Mutually exclusive--every unit fits into only
      one category.
     b. Equivalent--all categories are similar in
      type.
     c. Exhaustive--every unit fits into one of the
      categories.
     d. Any “other” category shouldn’t contain
      more than 10% of the units.
CONTENT ANALYSIS
   7. Categories also must be valid:
     a. Face validity
     b. Semantic validity
       1) Confirmation that words chosen for both units
        & categories mean what you say they mean.
       2) Panel of informed judges can provide
        confirmation.
     c. Criterion-related validity
     d. Construct validity
CONTENT ANALYSIS
   E. Coding the data
     1. Place units into categories (called coding).
     2. Researchers seldom (or rarely) code units
      themselves.
        a. Introduce bias or error into the analysis.
        b. Independent coders permit reliability to be
         assessed.
     3. Use at least two or more independent coders
      (also known as raters, or judges).
     4. Can also use a computer to do the coding.
CONTENT ANALYSIS
   5. Coders must be trained, but ideally are “naïve.”
   6. Do pilot study to obtain an interrater (intercoder,
    interobserver) reliability coefficient.
      a. Assess level of agreement among raters.

      b. If closer to 1.00, the greater the degree of
       intersubjective agreement (hence reliability).
   7. Revise coding scheme.
   8. Recode the data.
CONTENT ANALYSIS
   F. Analyzing & interpreting the data.
     1. After units numerically coded, apply statistical
       procedures to the data.
        a. Usually nominal or ordinal level data .
           1) Use a Chi-Square to assess differences or
            relationships.
           2) Can also determine means, modes,
            medians, etc.
           3) Can also do qualitative analysis.
CONTENT ANALYSIS
   b.   For interval level data, can also use
     inferential statistics:
       1) Use a t-test for comparisons.
       2) Use a frequency distribution.
  2. From analysis, interpret the data.
   a. Discuss relationship of the variables.
   b. If descriptive, need to justify
     interpretations.
III. Assessing Content-Analysis

  A.  Advantages
    1. Unobtrusive technique.
    2. Accepts unstructured material.
    3. Can handle large amounts of data.
    4. Studies communication in context.
    5. Can study changes in messages over time.
    6. Can be used to support theory.
    7. Easy to triangulate with both quantitative &
     qualitative methods.
ASSESSING CONTENT ANALYSIS
   B. Disadvantages
     1. Not relevant to all research (e.g. small numbers
      of texts).
     2. Cannot determine the truth of an assertion .
     3. Cannot evaluate the aesthetic qualities of a
      message.
     4. Establishing cause is tricky.
     5. Interpretations of meaning can be biased.
     6. Requires trained coders, so is time-consuming.
     7. If don’t use trained coders, reliability is suspect.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:12/24/2012
language:Unknown
pages:23