Page updated 11/18/09

FALL 2009


Frequently asked questions and answers
:

1. Could you give me an example of Type I and Type II errors?

                 A type I error occurs when a researcher rejects the null hypothesis when they should not.  This means that they are telling people that the use of a new counseling technique will improve self-esteem when it has not effect.  This is not good, counselors are better off using the usual techniques to improve self-esteem.  The researcher thus loses credibility.
                Another way to think about a Type I error is

                A type II error occurs when the researcher accepts the null hypotheses when in fact they should not.  This means that they are telling people that use of the new method of teaching reading does not produce better results than the usual method of teaching reading.  This is not as harmful as a Type I error since we are telling people to continue to do what they have done.

 2. Is statistical regression and multiple regression the same thing?

                 No, statistical regression is the tendency of subjects that have extremely high test scores to get slightly lower score the next time they take a test on the same subject and for  the tendency of subjects that have extremely low test scores to get a slightly higher score the next time they take a test on the same subject.

                Multiple Regression is a statistical test used to predict scores using a variety of variables.  For example, can we predict one counselor license exam score from variables such as one’s graduate GPA, GRE score, and 3 of hours in supervised counseling.

3. What statistical test should be used to determine if the differences between average scores for two groups?

   a t-test or an ANOVA


Sample test questions, answers will appear here later this week:

1. A research hypothesis should be stated in the form of a

a.   value judgment.

b.   null hypothesis.

c.   research question.

d.   predictive statement.

The answer is d.


2.
Which of the following variables is probably the most difficult to define operationally?

a.   classroom atmosphere

b.   performance in chemistry

c.
   score on a math test

d.   hand-eye coordination

The answer is a.  The other options are much easier to measure.

3. Which of the following is the best estimate of the standard deviation for this set of data: 10, 10, 10, 8, 8, 5, 4, 4, 4, 4, 2, 2, 1

a. 12        b. 9.87        c.  3.8        d. 127.5        e. .05

The answer is c.


A sample test item has been placed in LiveText for this course.  If you do not have a LiveText account you will need to acquire one soon.

APA citations:

“Within a paragraph, you need not include the year in subsequent references to a study as long as the study cannot be confused with other studies cited in the article.

When a work has two authors, always cite both names every time the reference occurs in the text.

When a work has three, four, or five authors, cite all authors the first time the reference occurs; in subsequent citations, include only the surname of the first author followed by et al. (not italicized and with a period after “al”) and the year if it is the first citation of the references within the paragraph.

When a work has six or more authors, cite only the surname of the first author followed by et al. (not italicized and with a period after “al”) and the year for the first and subsequent citations.” 

 


1. Please note: Sam has picked this research proposal topic: What is the relationship between self-esteem and algebra report card grades for 8th grade students? Sam may only be able to find one article on this exact topic.  If that is the case, it will be OK for him to find other related articles such as self esteem and math grades for middle or high school students; OR self-esteem and math test scores; OR studies that focus on factors related to success in algebra instruction.

2. If you have a question about LiveText please contact Katie at
           
thompsonk5@winthrop.edu

 

Vassar Stats

This webpage is useful to learn more about statistical procedures as well as run statistical analysis on small data sets.  The webpage was built and is maintained by Richard Lowry, a Professor of psychology at Vassar College.  The webpage can be found at:        http://faculty.vassar.edu/lowry/VassarStats.html

 

t
Announcements for College of Education graduate students are found at this site:

http://coe.winthrop.edu/graduate/announce.htm

 

 
Vassar Stats:

      http://faculty.vassar.edu/lowry/VassarStats.html

        
 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Research Proposal HINTS:

Summary

 At the least, your summary of each article should include the following:

 Where the research took place

 Subjects – number and ethnicity

 Key question of interest

 Critical components of the methods used in the research

            (how data was collected)

 Results

Critique

    You must critique the research methods.  You should not comment on the writing style of the author. A critique should include both positive and negative points about the research.  Some questions to think about:

    Was the sample large enough?
    Was the research conducted in a real school or health center?
    Did the sample include people from a variety of backgrounds?
    Did the research use valid and reliable instruments to measure the independent and dependent variables?

 
Sample APA references


Dewalt, M., & Rodwell, F. (1987). Effects of increased learning time in remedial math and science.
         ERS Spectrum, 6
(1), 33-36.

Dewalt, M., & Troxell, B. (1989).  Old Order Mennonite one-room school: A case study.    
        Anthropology and Educational Quarterly,
20(4), 308-325.

Dewalt, M., Vare, J. & Dockery, E. (2001). Retaining teacher candidates in the new

American college. (ERIC Document Reproduction Service ED455197)

Hostetler, J. (Ed.). (1989). Amish roots: A treasury of history, wisdom, and lore.
         Baltimore: Johns Hopkins University Press.

 
Research Proposal

    Each summary should include the following information:
        Key question of interest
        Sample selection, size and ethnicity
        Location of the project
        Method (how was the research conducted)
            also include key instruments used to measure independent and dependent variables
        Results - what were the key findings
    Each critique should include the positive and negative aspects of the research method

Some Topic Ideas

 

What is the relationship between pet therapy and depression in people over 65?

What is the relationship between the hours of tv watched per week and attention span of 3rd grade students?

What is the relationship between adding 5 minutes of sprinting to each half hour run and short term memory?

What is the relationship between walking six times per week for 30 minutes each session and self-esteem?

What is the relationship between adding six 30 second sprints to each distance run and time in the 10K run?

What is the relationship between self-esteem score and reading achievement for elementary age students?

What is the relationship between walking 3 times per week and self-esteem?

What is the relationship between group therapy and self-esteem of COA’s

What is the relationship between participation in after school programs and delinquent behavior?

What is the relationship between age of first tobacco use and marijuana use as an adult?

What is the relationship between anxiety and SAT scores?

What is the relationship between birth order and SAT scores?

What is the relationship between SAT scores and GPA for college freshman?

What is the relationship between type of therapy (group vs. individual) and self-esteem?

What is the relationship between the amount of alcohol consumed per week and weight?

What is the relationship between eating half a cup of oatmeal each day and cholesterol score?

What is the relationship between weight training and bone density for people over 65 years of age?

What is the relationship between reading ability and math test scores?

What is the relationship between amount of TV viewed per week and academic achievement?

What is the relationship between student time on task behavior and reading comprehension?

What is the relationship between birth order and reading achievement?

What is the relationship between amount of electronic entertainment and creative writing ability?

What is the relationship between 15 minutes of uninterrupted sustained silent reading (USSR) per school day and reading comprehension?

What is the relationship between type of art program and self esteem for middle school students?

What is the relationship between the amount of math courses taken in high school and freshman GPA?


Understanding Research Methods:  An Overview of the Essential
Selected Textbook Answers



Topic 9:  Quantitative vs. Qualitative Research: I
1. Quantitative
2. Qualitative
3. Quantitative
4. Qualitative
5. Qualitative
6. Quantitative
7. Qualitative
8. Quantitative

Topic 10: Quantitative vs. Qualitative Research: II
1. B
2. A
3. Qualitative
4. Qualitative
5. Qualitative
6. Quantitative

Topic 11:  Program Evaluation
1. Program evaluation
2. Program evaluation
3. Yes
4. Summative
5. Formative. (Note:  Measurement of job placement would be necessary to conduct a summative evaluation.)
6. Formative
7. Formative
8. Formative
9. Summative

Topic 35: Measures of Optimum Performance
1. Intelligence test
2. Aptitude test
3. Achievement test
4. Aptitude test
5. Checklist
6. By making sure that scorers know specifically what characteristics of the essays, products, or performances they are to consider and how much weight to give to each characteristic
7. No
8. They have low to moderate validity, with validity coefficients of up to about .60

Topic 36:  Measures of Typical Performance
1. No
2. To reduce the influence of social desirability in responses
3. The influence of social desirability
4. Projective techniques
5. Projective techniques
6. Strongly agree to Strongly disagree
7. Yes

Topic 37:  True Experimental Designs
1. To explore cause-and-effect relationships
2. By assigning participants at random to groups
3. Observation or measurement
4. Experimental treatment
5. Pretest sensitization (i.e., the interaction of testing and the treatment), which is also known as the reactive effect of testing
6. Posttest-only randomized control group design
7. There must be a reasonably large pool of participants to begin with so that when they are divided into four groups, each of the groups will have a sufficient number to yield reliable results
8. Random assignment

Topic 38: Threats to Internal Validity
1. Testing
2. History
3. Instrumentation
4. Selection
5. Maturation
6. When participants are selected on the basis of their extreme scores
7. By using a true experimental design

Topic 39: Threats to External Validity
1. External validity
2. Internal validity
3. Selection bias
4. Multiple-treatment interference
5. Reactive effects of experimental arrangements
6. Reactive effects of testing or pretest sensitization

Topic 40: Pre-Experimental Designs
1. No
2. No
3. Static-group comparison design
4. One-group pretest-posttest design
5. No

Topic 41: Quasi-Experimental Designs
1. Nonequivalent control group design
2. The dashed line
3. No
4. The same individuals are serving as both the experimental and control groups. Thus, they are the same in all ways at the beginning of the experiment
5. Multiple-treatment interference
6. A single-group design in which treatments are alternated

Topic 42: Confounding in Experiments
1. The explanation for a given difference
2. John Henry effect
3. The Hawthorne effect
4. An “attention-control group”
5. The tendency of individuals to improve (or at least feel that they are improving) simply because they know they are being treated
6. Double-blind experiment
7. The expected outcome of an experiment

Topic 43: Descriptive and Inferential Statistics
1. Descriptive
2. Nine
3. Percentage
4. Inferential
5. No
6. Inferential
7. Inferential
8. Parameters
9. Statistics

Topic 44: Introduction to the Null Hypothesis
1. Three
2. The observed difference was created by sampling error
3. Random errors
4. Zero
5. The null hypothesis
6. The probability that the null hypothesis is true is less than 5 in 100
7. When it is low
8. Fail to reject the null hypothesis
9. By saying that a difference is statistically significant
10. Dissertation

Topic 45: Scales of Measurement
1. Nominal
2. Interval and ration
3. Ordinal
4. Ratio
5. Nominal
6. Interval
7. Ordinal
8. Ratio
9. Ratio

Topic 46: Descriptions of Nominal Data
1. 60%
2. Yes
3. Bivariate
4. 100
5. One
 6 Percentages
7. Convert them mentally to percentages

Topic 47: Introduction to the Chi-Square Test
1. Observed results
2. (Random) sampling errors
3. Chi-square
4. No
5. p
6. To reject it
7. p<.05
8. No
9. No

Topic 48: A Closer Look at the Chi-Square Test
1. Bivariate analysis
2. Reject it
3. Five
4. 1 in 100 or less
5. Type II
6. Type I
7. Because it is free from bias

Topic 49: Shapes of Distributions
1. 60 (accept as correct an approximation of this value from the figure)
2. Vertical
3. The normal (bell-shaped) curve
4. Positive (skewed to the right)
5. Negative
6. Positive (skewed to the right)
7. Positive

Topic 50: The Mean, Median, and Mode
1.Mode
2. Mean
3. Median
4. The value around which the deviations sum to zero
5. Summing (adding up) the scores and dividing by the number of scores
6. No
7. Yes
8. Mode
9. Measures of central tendency

Topic 51: The Mean and Standard Deviation
1. Mean
2. The amount by which the participants vary or differ from each other (or differ from the mean)
3. Yes
4. Zero
5. 68%
6. 85 and 115. (Note: The mean of 100 plus/minus the standard deviation of 15 yields the answer.)
7. 34%
8. No
9. Group X
10. Group X

Topic 53: The Pearson Correlation Coefficient
1. Pearson product-moment correlation coefficient
2. -1.00
3. Yes
4. Yes
5. Yes
6. Direct
7. A perfect positive or direct relationship
8. 36
9. Multiply it by 100%
10. 49%. (Note: .70 x .70 =  49%.)

Topic 54: The t Test

1. There is no true difference between the means; the difference was created by random sampling
2. No
3. No
4. Yes
5. Yes
6. More
7. More
8. Yes
9. Yes

Topic 55: one-Way Analysis of Variance
1. Analysis of variance
2. Yes
3. Yes
4. No
5. No, because a t test is designed to compare only two means
6. Yes
7. No
8. p

Topic 56: Two-Way Analysis of Variance
1. One-way, because the participants were classified in only one way:  residence in urban, suburban, and rural areas
2. Yes. (Note: Praise rewards are more effective for older adults while monetary rewards are more effective for young adults.).
3. No. (Note: The column means are equal.)
4. No. (Note: Method A is 15 points superior to Method B regardless of aptitude level.)
5. Yes. (Note: Method A has a higher column mean.)
6. No. (Note: The row means are equal.)
7. No

Topic 57: Practical Significance of Results
1. Yes
2. Five
3. Yes
4. No
5. Yes
6. False
7. No

Topic 58: Introduction to Effect Size (d)
1. Experimenter A, whose scores could range from only 0 to 20. (Note that Experimenter B’s could range from 0 to 120.)
2. Their standard deviations
3. Yes. (Note that this is the case in the example in this topic.)
4. Yes, it is a large difference because it is close to 3.00, which is the effective top of the range for values of d because there are, for all practical purposes, only three standard-deviation units above the mean
5. No, because the values of the standard deviations must be known to calculate d
6. 1.00. (Note: 500.00 – 400.00 = 100.00/100.00 = 1.00.)
7. Effect size refers to the magnitude (i.e., size) of a difference when it is expressed on a standardized scale.

Topic 59: Interpretation of Effect Size (d)
1. No
2. Three.  It can range three standard deviations above and three standard deviations below the mean.  Less than one-half of one percent of a distribution lies above +3.00 and below -3.00
3. One standard-deviation unit
4. 1.40 or more (i.e., 1.40+)
5. Large
6. When the control group’s mean is higher than the experimental group’s mean
7. Before

Topic 60: Effect Size and Correlation (r)
1. Cohen’s d and “effect size r” (i.e., the Pearson correlation coefficient [r] and its associated value of r2
2. +1.00
3. Squaring r and multiplying it by 100%
4. 16%. (Note: 0.40 x 0.40 = 0.16 x 100% = 16%.)
5. r and its equivalent r2
6. 0.264. (Note that the answer is obtained from Table 1.)
7. 1.50. (Note that the answer is obtained from Table 1.)

Topic 61: Introduction to Meta Analysis
1. Occurring later and/or being later and more highly organized. (Note: See Footnote 1 in the topic.)
2. Combining the results of previous studies
3. Random sampling errors created by random assignment and random errors of measurement
4. Mathematical synthesis of the statistical results of previous studies
5. It typically synthesizes the results of studies conducted by independent researchers

Topic 62: Meta-Analysis and Effect Size
1. No
2. Various researchers frequently use a number of different measures of the same variable
3. Use a measure of effect size
4. Studies with larger samples need to be given more weight than studies with smaller samples
5. No, because r is also used

Topic 63: Meta-Analysis: Strengths and Weaknesses
1. 109,654
2. A meta-analysis
3. No. A set of studies subjected to meta-analysis might all have a common flaw that biases (and makes invalid) the combined results from all of them
4. Publication bias
5. A partial solution to this problem is for those conducting meta-analyses to search for studies that might be reported in dissertations, convention papers, government reports, and other nonjournal sources for inclusion in their meta-analyses

Topic 64: Sampling in Qualitative Research: I
1. Qualitative researchers
2. Their ability to provide the needed information
3. Purposive criterion sampling
4. No
5. No

Topic 65: Sampling in Qualitative Research: II
1. No
2. A broader interest than just a single source (such as a single college campus)
3. By renting lists from professional associations or list brokers
4. 13
5. Qualitative researchers
6. At the point at which several additional participants fail to respond with new information that leads to the identification of additional themes

Topic 66: Measurement in qualitative Research: I
1. Yes
2. Written directions for conducting the interview as well as a standard set of predetermined questions to be asked of all participants
3.At least a few individuals who will not be participants in the study
4. Establish rapport
5. Yes
6. Self-disclosure

Topic 67: Measurement in Qualitative Research: II
1. Phenomenological approach
2. Six to twelve
3. Facilitator and moderator
4. Focus groups reveal the evolution of perceptions in a social setting
5. Nonparticipant observation
6. Ethnography

Topic 68: Quality Control in Qualitative Research
1. Methods triangulation
2. Two or more
3. Researcher triangulation
4. Yes
5. An outside expert; a research’s peer such as another experienced qualitative researcher
6 The participants

Topic 69: Grounded Theory for Data Analysis
1. No
2. Inductive approach
3. Open coding
4. Axial coding
5. Axial coding
6. Constant comparison

Topic 70: Consensual Qualitative Research: Analysis
1. False
2. Code into domains
3. Developing core ideas within domains
4. Core ideas are grouped into categories based on similarities
5. General
6. No




APA INFO:

If you quote material from an article, APA style requires you to list the page number where the quote can be found in the article.  If you have an online article with no page numbers then you would list the paragraph number.

Example of APA citation for a quote:

Dewalt (2006, p. 2) stated, "In Pennsylvania, Amish parochial schools are generally one-teacher schools."

OR

"Students in the study completed the written survey during a free period" (Conley, et al., 2006, p. 14).

Other citations in text:

APA manual page 208 & 209 - if a work has 6 or more authors you only cite the first author in the paragraph

Wonderlich et al. (2001) conducted a study

APA manual page 240 - In the reference section you will list the first six authors and then et al.

APA Info: Personal Communications 

Personal communications may be letters, memos, some electronic communications (e.g., e-mail or messages from nonarchived discussion groups or electronic bulletin boards), personal interviews, telephone conversations, and the like. Because they do not provide recoverable data, personal communications are not included in the reference list. Cite personal communications in text only. Give the initials as well as the surname of the communicator, and provide as exact a date as possible:

             B. Smith (personal communication, April 18, 2007)

            (V. White, personal communication, September 28, 2006)



APA info:

When an article has one or two authors you will state both names every time you cite that article in your writing.

If an article has 3 - 5 authors you will only state all authors the first time you cite the article. In subsequent citations of the article you will state the first author's name and et al.

If an article has 6 or more authors you will only state the name of the first author and et al. when citing the article.

Check this link for more information relative to graduate students in the COE:
  http://coe.winthrop.edu/graduate/announce.htm


The Statistics Homepage: http://www.statsoft.com/textbook/stathome.html 

Research Proposal:  The critique section should focus on the strengths and weaknesses of the research methods.  Did the project have a good sample?  Was the sample large enough?  Did they use reliable or valid tests? 

        Citation of a work discussed in a secondary source

Text citation:

Fisher and Stahl’s study (as cited in Dewalt and Troxell, 1989) found that Amish children spend two hours per day on individual study.

Reference Page:

Reference page will include Dewalt and Troxell but not Fisher and Stahl.


Suggested web page: http://faculty.vassar.edu/lowry/VassarStats.html


Dr. Dewalt's Home Page
 


                                                                  Hit Counter