4130 Chapter 11: Non experimental designs

Card Set Information

4130 Chapter 11: Non experimental designs
2012-06-22 18:45:25

Non Experimental Designs
Show Answers:

  1. Chapter 11: Non Experimental Designs
  2. Nonexperimental design rationale
    **relationships or differences among variables 
    *Also called an observational study.

    Used in studies to:

    •Construct a picture of a phenomenon at  one point or over a period of time.

    •Explore people, places, events, or situations as they naturally occur.

    •Test relationships & differences amongst variables.
  3. Non experimental design
    *The IV is NOT manipulated but occurs naturally.

    *The researcher looks at relationships or differences between variables.

    *The researcher requires a clear, concise research problem or hypothesis that is based in a theoretical framework. (Remember a theoretical framework is theory that has been tested.)

    *It is on the lower end of the research continuum: level 4 evidence

  4. Types of nonexperimental research designs
    1. Survey studies: descriptive, exploratory, comparative.

    2. Relationship or difference studies: correlational, developmental (cross-sectional, longitudinal or prospective, retrospective or ex post facto). 
  5. 1. Survey
    when is this used?
    exploratory/descriptive & comparative surveys 
    advantages and disadvantages 
    Used when the researcher wants to assess the nature of the problem being studied.

    *Observations are made instead of intervening by the researcher.

    *When little is known about variables.

    *Exploratory or descriptive used to search for accurate information about the characteristics of particular subjects, group, situations or frequency of a phenomenon.

    *Comparative surveys when they are used to determine differences between variables.  IVs are not manipulated but data is assessed to provide evidence for future nursing interventions.

    • Advantages & Disadvantages:
    • Advantages
    • They provide a great deal of accurate information about the larger population that is collected in an economical manner (for the amount of information gathered)

    • Disadvantages
    • They tend to be superficial gathering breadth rather than depth of information. They can be time consuming & costly. 

  6. 2. Relationship or difference studies
    pearson r 
    advantages and disadvantages
    developmental studies: advantages & disadvantages
    *As the title notes this type of study traces the relationships or differences between variables to gain a greater depth or understanding about phenomenon = correlational & developmental studies

    • A.Correlational studies: looks at relationships. The researcher looks to see if one variable changes how the other variable changes. Looking at quantifying the strength of the relationship between variables or in testing a hypothesis about a specific
    • relationship. The relationship can be positive or negative.

    This is done with the Pearson r correlation coefficient.

    Pearson r

    *This is a statistical test that is calculated to reflect the degree of relationship between 2 variables (p. 355).

    *Once the r is calculated the researcher reports the value of the correlation & the probability of it occurring by chance.

    *Correlation coefficients range in value from -1.0 to +1.0. The value in between of 0 indicates that there is NO relationship between the variables.

    *+1 is a perfect positive relationship. -1 is a perfect negative relationship.

    *Example: I am interested to know the correlation between the age of a patient & the time required to recover from surgery.

    • *A Pearson correlation test shows that  r = +.76 at p = 0.025. From this I know that there is strong positive relationship or if
    • one is older the more recovery days. The low p values tells me that my result is statistically significant or the probability of the test result occurring by chance is  very low or the probability of the negative strong negative relationship is high. The values of the r & p are considered together.


    *↑flexibility when investigating complex relationships between variables.

    *Efficient & effective method of collecting a large amount of data.

    *Potential for practical application in clinical settings.

    *A potential foundation for future experimental studies.

    *A framework for exploring the relationship between variables that cannot be manipulated. 


    *The variables cannot be manipulated.

    *Randomization is not used thus generalizability is decreased.

    *Casual relationships cannot be made because of reduced controls, manipulation or randomization.

    B. Developmental studies

    These types of studies use a time perspective or the changes that occur from elapsed time.

    1. Cross-sectional studies: data is collected on participants at 1 point in time. Less costly & time consuming than longitudinal studies. Not affected by maturation.

    2. Longitudinal or prospective studies (repeated measures): data collected from the same group at different points in time. They are costly in regards to time, money & effort. Hawthorne effect & maturation are threats. They begin in the present & end in the future.

    3. Retrospective or ex post facto (after the fact) studies: the DV has already been affected by the IV. The researcher attempts to link present events to events that have occurred in the past. See model below.


    *Important designs when randomization, control & manipulation are not appropriate or possible.

    *Useful in testing theoretical models of how variables work together in a group in a particular situation.

    *Important to develop knowledge base on phenomenon of interest.

    *Useful in forecasting or making predictions.


    *Difficulty explaining cause-and-effect or causal relationships.

    *Challenging to find groups that are similar in extraneous variables (ex facto studies). 

  7. Methodological Research
    *Used to develop & evaluate data collection instruments, scales, and techniques.

    *Psychometrics = theory and development of measurement instruments (surveys or questionnaires) & measurement techniques. Most nursing concepts have some sort of measurement tool already developed such as pain, stress, quality of life, knowledge sharing, etc.

    *The tools are tested to ensure they are:

    *Reliable: if a variety of people used the same tool they would generally get the same answers.

    *Valid: they measure what they are meant to measure.

    *This study method includes defining a concept, formulating the items on a tool, developing instructions for user of the tool & testing the tools reliability & validity. 

    • _____________
    • *A meta-analysis is a strict scientific process that synthesizes the findings from several separate studies in a specific area & statistically summarizes the findings to obtain a precise measure of the effect.

    *The Cochrane Data Base houses only meta-analysis and systematic reviews of literature.

    *A secondary analysis is a form of research in which the researcher takes previously collected data from a previous study & reanalyze the data for a secondary purpose. Public data bases can be used to take information from & the researcher performs statistical tests on it. 
  8. Nonexperimental Critiquing Criteria
    *Which nonexperimental design is used?

    • *Is the rationale for this design evident such as a
    • theoretical framework?

    *Is the design congruent with the purpose?

    *Is the design appropriate for the research problem?

    *Is the design suited to the data collection method?

    *Are the findings presented appropriately?

    *Is cause-and-effect inferred?

    *Are alternative explanations addressed?

    *Are the threats to validity addressed?

    *Are the limitations addressed?
  9. Review
  10. This type of design is better known for the breadth of data collected than for the depth of its data
  11. a moajor disadvantage of this design is the length of time needed for data collection
  12. the main question is wheter or not variables covary
  13. these words mean "after the fact"
    ex post facto
  14. eliminates the confounding variable of maturation
  15. quantifies the magnitude and direction of a relationship
  16. collects data from the same group at several points in time
  17. this can be surprisingly accurate if the sample is representative
  18. uses one data from one point in time
  19. based on 2 or more naturally occurring groups with different conditions of the presumed independent variable
    cross sectional