september 19, 2006 cp 6002 statistics and research ii

26
September 19, 2006 CP 6002 Statistics and Research II

Upload: mildred-garrison

Post on 26-Dec-2015

213 views

Category:

Documents


0 download

TRANSCRIPT

September 19, 2006

CP 6002 Statistics and

Research II

Can you name some ‘things’ that change or has the potential to change (in quantity)?

Can you name some ‘things’ that are unchanging?

Variables vs. Constant

Variable = any characteristic or quality that VARIES among the members of a particular group

Constant = any characteristic or quality that is the SAME for all members of a particular group

Types of Variables Quantitative Variable = a variable that varies in

amount or degree, but not in kind Categorical Variable = a variable that varies only in

kind, not in degree or amount Independent Variable = a variable presumed to

affect or influence other variable Dependent Variable (outcome variable) = a

variable presumed to be affected by one or more independent variables

Extraneous/Confounding variable = an independent variable that may have unintended effects on a dependent variable in a particular study

Hypotheses

A prediction of results made before a study commences (statement of what we expect)

Researcher’s tentative prediction (reasoned - not a wild guess!)

Leads to a greater amount of important knowledge (if it is a significant and correctly stated hypothesis)

Types of Hypotheses Directional = a prediction about a specific nature of

relationship (or difference); e.g. method A is more effective than method B

Non-Directional = a prediction that a relationship (or difference) exists without specifying the exact nature (direction); e.g. there will be a difference between method A and method B (without saying which will be more effective)

Null = states that ‘there will be no significant relationship (or difference) between variables (presumed to be true until statistical evidence indicates otherwise!)

Reminder…

Researchers do not set out to ‘prove’ a hypothesis

Rather, they collect data that either support or do not support it (hypothesis testing)

A hypothesis is formulated based on a theory (deductive hypothesis) or the review of related literature (inductive hypothesis)

A Good Research Hypothesis

Sets up a ‘testable’ situation

Gives direction to research

Identifies the variables of importance

Is grounded in theory

Is brief but with clarity

Exception

Some studies use objectives or questions instead of hypothesis – descriptive, ethnography

E.g. a descriptive study of counselors’ salary schedule and policies

Objective: DESCRIBE level of salary for state and educational levels

Question: What are the kinds of salary schedules in place for counselors…?

What are some of the dos’ and don’ts of Writing a Review of Related

Literature?

An Example

Sexual harassment has many consequences. Adams, Kottke, and Padgitt (1983) found that some women students said they avoided taking a class or working with certain professors because of the risk of harassment. They also found that men and women students reacted differently. Their research was a survey of 1,000 men and women graduate and undergraduate students. Benson and Thomson's study in Social Problems (1982) lists many problems created by sexual harassment. In their excellent book, The Lecherous Professor, Dziech and Weiner (1990) give a long list of difficulties that victims have suffered.

Another Example

The victims of sexual harassment suffer a range of consequences, from lowered self-esteem and loss of self-confidence to withdrawal from social interaction, changed career goals, and depression (Adams, Kottke, and Padgitt, 1983; Benson and Thomson, 1982; Dziech and Weiner, 1990).For example, Adams, Kottke, and Padgitt (1983) noted that 13 percent of women students said they avoided taking a class or working with certain professors because of the risk of harassment.

What’s the difference between the two? Which one do you think is a good

RRL?

What RRL is not?

NOT a descriptive list of the information gathered

NOT summary of one piece of literature after another

Review of Related Literature

To demonstrate familiarity with a body of knowledge

Establish credibility – increase readers’ confidence Show the path of prior research Show how current research is linked to past work Integrate and summarize – pull together and

synthesize different results Point out areas where prior studies agree, where

they disagree, and where major questions remain Learn from others and stimulate new ideas

Writing a good RRL

Remember the purpose

Read with a purpose

Write with a purpose

Work toward a specific focus

The review of literature should be like a discussion with a friend concerning the studies, research reports, and writings that bear directly on your own effort. Be very clear in your thinking.

Organize a plan

Have an outline (guiding concept), for which the best guide is the problem itself.

Begin the discussion from a broad perspective and narrow to the specific problem.

Stress Relatedness (relevance)

Remind the reader constantly of how the literature you are discussing is related to the problem.

Review the Literature, Don’t Copy it!

Review the Literature; Don’t Copy It!

More important than what the study says is what you say about the study.

Summarize your thoughts and ideas

Continue asking the question, "What does it all mean?" and continue searching for relatedness.

Keep your own voice!

Be careful with paraphrasing

When to stop?

Circular pattern (same arguments, same findings, etc.)

Rough draft (revise, revise, revise)

Others’ feedback

Models

Old and new references

The Process

Secondary Sources

General References (Virtual Library, EBSCO, PsycINFO, PsycARTICLES,

etc.)

Primary Sources

Formulating Search Terms

Obtaining and Reading the Primary Sources

Note Cards

Problem (topic) Hypothesis Participants Procedures Findings Conclusions

Work out

1. the comparison the writer establishes in the review

2. the sequence to his review (why that sequence?)

3. what the writer's own perspective is

Take note of…

1. the use the writer makes of each of the sources he refers to

2. how, in his language particularly, he avoids a "black and white", right/wrong type of judgment of the positions he reviews