Chapter 13 Qualitative Data Analysis

Document Sample
Chapter 13 Qualitative Data Analysis Powered By Docstoc
					Ethics and Qualitative Analysis
Objectives: Describe and illustrate the ethical issues involved in research Describe the role of the IRB Differentiate between political and ethical concerns Identify why politics are important Access and use the NC State IRB format Illustrate how patterns can be found Describe the stages of the constant comparison technique used with the grounded theory method Show how coding works in qualitative analysis Define open, axial, and selective coding Summarize the role of memoing and notes-on-notes Explain the role of conceptual mapping Appreciate how computers can be used in qualitative data analysis

Mixed Methods, Linking, or Triangulation (Left overs from 3/11)


 


Linking—using more than one data source, method, theoretical framework, or type of data Has both advantages and disadvantages Advantages: gives more information, provides a reliability and validity check Disadvantages: takes time and money, requires rigor for all parts, data may not agree

Linking


     

More than one source More than one method More than one theoretical framework More than one type of data (More than one researcher/evaluator)

*Usually operates out of Positivist Framework *Kinda like a case study

Linkage Techniques
Antecedent or sequential  Encapsulated or nested  Concurrent  Primary/secondary contributions




Can also be ―within-method‖

Linking Issues


   


Complementariness Congruence-similarity, consistency, convergence Divergence  MUST HAVE THE EXPERTISE to do this Time consuming and costly—better to do one thing well than two things poorly!!! Linking does not automatically ensure trustworthiness

Evaluation/Research Review
   

  

Identify the focus of the investigation Outline what needs to be studied Select appropriate measurement tools Develop a plan to collect data Collate all data Interpret data Make recommendations from the study

Broad Issues
Ethical  Political  Legal  Moral


Researcher/Evaluator must be Competent
   



Knowledge about area to be researched Knows about research design Knows wide range of research methods Knows how to analyze and interpret data and relate them back to theory Know how to analyze both qualitative and quantitative data and when best to use each

Developing Competencies – con’t
Knows how to use the results  Understands how to handle political, legal, and ethical concerns encountered  Must have certain personal qualities: trustworthy, strives for improvement, responsive to sensitive issues


―Doing the Right Thing‖


Political Issues (science is touched by politics but goes on anyway; social change is always political; values matter)
  

 

Supports/refutes views and values Personal contacts Value-laden definitions Controversial findings Pressures to produce certain findings



Know the organization‘s position, don‘t go beyond the data in conclusions, have clear purpose for research

―Doing the Right Thing‖


Legal Issues


Not many legal concerns except around illegal behaviors



Moral
 

Unintentional mistakes made by bias or mistake Cultural & procedural biases  Letting bias, prejudice, friendships influence outcomes  Dealing with negative findings  Taking too long to get the results out  Be prepared to recognize the possibility of statistical errors and know how to explain them

―Doing the Right Thing‖
Voluntary Participation  No Harm  Anonymity and Confidentiality  Issues of Deception  Analysis and Reporting


Ethics Principles
 Be

open with people  Don’t promise more than you can do  Protect the rights and privacy of your subjects  Guard against coercion  Get written consent & Board approval  Guard against harm  Let participants know the results



Anonymity The researcher cannot identify a given response with a given respondent. Confidentiality Researcher can identify a given person's responses but promises not to do so publicly.





Informed consent Subjects in a study must base their voluntary participation on a full understanding of the possible risks involved.

How Politics and Ethics Differ
Both ideological  Ethics focuses on methods: Politics center on substance and use  No formal codes of political conduct  Use of statistics can be political (Best indicates)


Ethics and Statistics



   



Competence in numeracy Mutant statistics due to flawed definitions, poor measurements, or bad samples Transforming data Garbling data Compounding mutant statistics Inappropriate comparison: over time, projections (cause and effect), among places, among groups, among social problems KNOW ‗DEM APPLES AND ORANGES

Institutional Review Board
Guidelines:
 Minimize

the risk to subjects  Risk must be reasonable to benefits of the study  Informed consent  Adequate detail should be provided about the study in lay language  Privacy must be maintained  Freedom of choice, free to withdraw at any time

Three Categories


Exempt
 Instructional strategies

tests  Previously collected data  Public benefit programs  Consumer acceptance



 Educational

Expedited Full

IRB LINKS


Form - http://www.ncsu.edu/sparcs/forms/irb_app.doc



Main page – http://www.ncsu.edu/sparcs/IRB/
OHRP federal regulations http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr4 6.htm





Policies and procedures http://www.ncsu.edu/sparcs/IRB/pol_proc.doc

What would you do?




You are in the grad room and discover that a file containing all the names of interviewees is left open. You collect data from several classes for a research study you are doing. One of the students comes in and tells you that the students were having a contest in one class to determine who could provide the wildest responses.

What would you do?




You have collected IRB forms from your participants in your study, but somehow your roommate thinks they are old papers that need to be recycled. In a class you help your instructor collect data for a research project. A year later, you see that the instructor has published the paper with no acknowledgement of anything that the students did to assist.

What would you do?
You have written your thesis or a really outstanding project. Your advisor wants you to publish the paper but you don‘t want anything more to do with it.  You complete your research and discover that your data reflect very badly on the people you have studied.


What would you do?




In the process of collecting data for your study, you find out that the head of the sports organization you are studying is sleeping with one of the players. You are doing 10 interviews for your project. After you have completed, transcribed, and done most of your data analysis, one of the participants comes to you and says they do not want anything they said to be included as part of your research.

Qualitative Analysis Types
Enumeration  Grounded Theory Method (that uses constant comparison)


Transcribing and Organizing
Data don‘t exist unless they are written  Transcription-taking oral conversation to words  Organizing data for analysis  Computer programs to assist


Steps in Qualitative Analysis
Read, read, read  Data reduction (open coding)  Data refinement  Data display/notes (memos)  (Constant comparison)  Selective coding and theory  Writing (emic, etic, thick description, rival hypotheses)


Coding


What about coding?



  

 

Usually descriptive the first time through Write-ups (the important content) and ―notes on notes‖ (short summaries of interpretations, patterns, and clarifications) Be organized (computer software, matrix, visual maps) Must spend a lot of time with the data Document how you code (trustworthiness) Integrate eval questions (criteria), data, & theory Remember that the analysis EVOLVES!

Enumeration
Code and count technique (pro/con)  Use the number of counts, not %  Still use the words (or quotes) to help illustrate and interpret the counts


Enumeration Example


What are you looking forward to next year the most?

   





  

The people Physical activities All camp and evening activities Having fun Sleeping/staying up later Everything New games The change Being myself Focus Day Nature

13 8 6 4 2 2 2 1 1 1 1

Coding Practice
Determine codes  Do counts




Grounded Theory Method (GMT) An inductive approach to research in which theories are generated solely from an examination of data rather than being derived deductively.



Constant comparative method Component of the Grounded Theory Method in which observations are compared with one another and with the evolving inductive theory.



Concept Mapping Putting concepts in a graphical format.

Computer Analysis
Extremely helpful in data organization and analysis  CAN‘T DO INTERPRETATION OR THEORY BUILDING  Common and good programs:

 NUD*IST  NVivo  Atlas

ti

Qualitative Analysis
Relation of data collection to data coding to analysis to interpretation to theory to writing  Analysis= process of bringing order to data by organizing into basic descriptive units, categories, and patterns  Interpretation= attach meanings & significance to analysis, explain patterns, develop theory


Coding
 Open

 Axial  Selective  Use

of Memos and Notes  Emic and Etic together (use of direct quotes=thick description)
 Remember that

the analysis EVOLVES!



Semiotics Study of signs and and the meanings associated with them.



Memoing Writing memos that become part of the data for analysis in qualitative research such as grounded theory.



Open coding Initial classification and labeling of concepts in qualitative data analysis. Codes are suggested by researchers examination and questioning of the data.

What to Code (Examples)
strict inclusion (X is a kind of Y)  spatial (X is a part of Y)  cause-effect (X is a result of Y)  rationale (X is a reason for doing Y)  function (X is used for Y)  sequence (X is a step in Y)  attribution (X is a characteristic of Y)


Memoing/Notes-on-notes
Notes to yourself  Five types:

 Code

notes  Theoretical notes  Operational  Sorting  Integrating

Constant Comparison
 

READ, READ, READ the data! 4 steps:
 Reduce,

code, display the major themes or patterns to emerge  Integrate categories- compare to each other and the themes  Delimit and refine the themes  Provide examples from data to show how themes derived (use quotes)

Constant comparison con‘t
Not looking for an ―average‖- rather, you want to describe the complete range within a theme (outlyers important and everything should fit in somewhere)  In research, searching for the ―new‖ slant from your data- emergent theory!  Content analysis is a form of constant comparison


Analysis Practice
Open Coding  Axial Coding  Notes/Memos  Thick Description  Theory Building


Data from Spring Break
Open Coding  Notes on Notes  Analyses Possibilities


Analysis Example


The purpose of this study was to explore what made for a memorable Spring Break experience for PRTM graduate students. The results represented three themes that could be described: relationships, typical and atypical activities, and emotional ranges.

Interpretations of Data


  

With qualitative data, analysis and interpretation occur simultaneously Your questions evolve and change as you code and analyze Have to be very familiar with data to do a good job of interpreting Use emic (ideas expressed by respondents) and etic (data expressed in evaluator‘s language) to provide thick descriptions

Interpretations con‘t
Can‘t make cause and effect type statements with qualitative analysis  Can‘t make comparisons of difference because we can‘t provide direct relationships


Emic and Etic


Students described relationships (or the lack of them) as important in memorable spring breaks The major relationships discussed were with family, friends, and new relationships. For example, one student described a trip to the ACA Tournament and how the ―fan support was amazing.‖ Several students described family trips for a cruise or for a skiing outing. One student noted, ― It was that week that defined the course of the relationship [with boyfriend]. Two months later I was married and the rest is history.‖ Friends were also evident in trips to places like the beach, mountains, or on cruises. Another student noted that, ―Spring break has always been reserved for family and friends.‖

More Emic and Etic


Responses regarding Spring Break included both typical and atypical activities associated with the ―time off.‖ Travel was an atypical activity from daily life such as visits to the ocean, the mountains, or to sporting events. Several students could be described as typical because the break did not involve something different or unusual. Along with the destinations was the opportunities at Spring Break that included food, drink, and the weather. One student remarked, ―The days had a similar pattern to them and that was getting up early, going to the beach, taking a nap, and then seeing the night life.‖ On the negative side, however, one student noted that ―I had planned to get away but thanks to a few things outside my control, that didn‘t work.‖

More…


All students described an emotional association with Spring Break in some way. These ranged from almost all positive such as experiencing new opportunities, relaxing, not worrying, and being happy to a sense of disappointment by a small minority of students. One student went to New Orleans and noted, ―This was my first trip to the Big Easy. We went to cool bars and a famous Thai restaurant where you sit on the floor.‖ Another student described a cruise and noted that ―the water was beautiful, warm, and clear.‖ Similarly another beach visitor remarked. ―…the whole week was spent without a care in the world.‖



More….

Interpretation (Theorizing) of Data




On means for analyzing these Spring Break data is to consider the aspects of how Spring Break parallels the ―recreation experience.‖ Clawson and Knetsch (1967) described a recreation experience as having anticipation, travel to, the actual activity, travel from, and recollection. These elements were evident in many of the experiences described by these PRTM students. Therefore, Spring Break is not just an event but an experience that cuts across the entire semester. This research also adds theoretically to the myriad of definitions used to describe a leisure experience when one has the opportunity to choose what he or she will do. Collectively the students of PRTM that were studied, generally found relaxation in a variety of activities. Spring Break cannot be defined as just one type of experience that makes it meaningful.

Ensuring Trustworthiness


   
   

Audit Trail (keep track of coding and thought processes) Prolonged engagement Range of cases Examples, quotes are data (Thick description) Research biases described Competence Major time commitment Triangulation/Linking Explain ALL variance

Writing Qualitative Data
Patterns/Themes  Data presentation  Emerging Theory

 Substantive

Theory  Formal Theory (Our SB DATA)

Instrument Draft- for March 25


 



Draft a 2-page questionnaire (with at least 10 questions) that you might use for something you are doing in research or other professional work Use at least two CLOSE-ENDED question structures and two types of question content Pilot test the questionnaire (with at least 2-3 people) and revise it Write a one page description of your questionnaire purpose and what changes you made in the questionnaire after it was pilot tested (you might include the original as well as your revised questionnaire)


				
DOCUMENT INFO