Research Tutorial

For the following data set, calculate the following ... Data: 15, 2, 23, 7, 18, 13, 9, 23, 8, 16, 27, 28, 12, 4, 28, 12, 18, 18, 14, 6, 15, 15, 31, 15, 15

1. What is the mode? 15 2. What is the median? 15 3. What is the mean? 15.68 4. What measure of central tendency would be the best estimate of this data? mean because there aren't any major outliers 5. What is the range? 29 6. What is the hinge? the median of the high and low portions of a data set 7. What are the hinge positions (upper & lower)? 12, 18 8. What is the midrange? 16.5 9. What is the low range? 13 10. What is the high range? 16 11. Compute the sample variance: s^2 = 59.393 12. Compute the sample deviation: s = 7.7

Which of the following is NOT an essential characteristic of a true experiment:

Careful measurement of the dependent variable

Suppose I want to see if review sessions help students improve their test scores in MSPT I. I get two groups of students. One group contains students who did best on test 1 and another group who did worse on test 1. If regression to the mean occurs, what will happen to the scores of members in the high & low ability groups on test 2? A. scores will stay the same B. scores from the group that did best will do better C. scores from the group that did worst will do worse D. scores from the group that did best will go down and those who did worst will do better

D

The experimenter can decrease the effects of extraneous variables by? A. randomization B. blocking C. random assignments D. all of the above

D

What is one important function of the debriefing or conclusions at the end of the experiment? A. to thank individuals for their participation B. to see if the individuals guessed the purpose of the study C. to correct any misperceptions the individuals had about the study D. all of the above

D

What's the purpose of randomization of subjects to groups? A. to control extraneous variables B. the same purpose as random sampling C. to make the groups exactly equal on an extraneous variable D. all of the above

D

Which of the followings an example of absolute or true research? A. measuring the self-concept of forty year old males B. examining the difference between two teaching methods C. comparing this year's high school graduates to last year's graduates D. examining the effects of two different levels of rewards on the run speeds of two groups of rats

D

What is the major wording problem with the following question: Do you think it is all right for underage people to drink alcoholic beverages, but not all right to drive a car after having several drinks?

Double-barreled question

All of the following are values for pre-testing questions on your survey except? A. determine if a question is unclear B. determine the time needed to complete the survey C. determine if the survey is too long and tedious for respondents D. all are values for pretesting questions

D

Determine IV, DV, & choose the appropriate statistical test: A group of students were asked to watch a violent TV show. Students were asked to fill out the measure of verbal aggression (VA) to assess their levels of VA prior to watching the TV show and then again after watching the TV show. Researchers want to determine if there is a different between these students' VA before and after the TV show.

IV: TV show DV: VA Dependent t-test (same sample)

Determine IV, DV, & choose the appropriate statistical test: Researchers wanted to know if certain age groups preferred certain movies. Therefore, researchers asked participants their age ~ 0-12 = children, 13-24 = adolescents, 25-36 = young adults; and then asked them which movie they preferred (horror, romance, comedy, action). The researcher coded children as 3, adolescents as 2, and young adults as 1.

IV: age groups DV: movie preference ANOVA

Determine IV, DV, & choose the appropriate statistical test: Females were asked to indicate what type of movie they preferred (Horror, romance, comedy, or action). Researchers wanted to know if females significantly differ on their preference of movies.

IV: females DV: type of movie Descriptive

Determine IV, DV, & choose the appropriate statistical test: A researcher wants to know the difference between males and females on their level of verbal aggression (verbal aggression is measured by Infante's 20 item Likest-type scale)

IV: gender DV: VA Independent t-test (2 diff samples)

Determine IV, DV, & choose the appropriate statistical test: Males and females were also asked to rate their preference of 10 TV shows from like the most to like the least. Then researchers want to know if men and women differed on their preference of TV shows.

IV: males & females DV: TV shows Chi-square

Determine IV, DV, & choose the appropriate statistical test: Researchers wanted to determine whether or not the medium of a story would influence their attitude towards the story. Therefore, researchers randomly assigned subjects into three groups: Group one: read the book "Tuesdays with Morrie" Group two: watched the movie "Tuesdays with Morrie" Group three: heard the audio version of the book "Tuesdays with Morrie" Then, each group was asked to fill out a 20 item semantic-differential scale to assess their attitude towards the story "Tuesdays with Morrie"

IV: medium DV: attitude toward story ANOVA

Determine IV, DV, & choose the appropriate statistical test: The researcher wanted to know if individuals in different relationship stages (single, casual dating, serious dating, and married) would differ on their relationship satisfaction & trust

IV: relationship stage DV: satisfaction/trust ANOVA

History is a threat to:

Internal validity

Temperature measured in degree Celsius is an example of what scale of measurement?

Interval

Nominal data levels of measurement

In a nominal level variable, values are grouped into categories that have no meaningful order. For example, gender and political affiliation are nominal level variables. Nominal data cannot be used to perform many statistical computations, such as mean and standard deviation, because such statistics do not have any meaning when used with nominal variables. However, the chi-square test can be performed on a cross-tabulation of nominal data.

You do not smoke, do you?

Leading question

What type of survey should be used to see if change in a variable occur over time?

Longitudinal

You randomly assign subjects to one of two rooms, a light green room or a light orange room, and want to determine the effects that room color has on test taking.

Room color = IV Test scores = DV IQ of students = extraneous variable

Descriptive statistics

Mean, mode, median to measure data

Response effects in survey represent a problem in ?

Measurement validity

Chi square test

Measures nominal & ordinal data

Dependent t-test (aka paired t-test)

Testing two variables using the SAME sample

Independent t-test (aka unpaired t-test)

Testing two variables using two DIFFERENT samples

Which experimental design controls for most threats to internal validity, therefore it provides the strongest evidence for causal relationships?

True experiment

Ordinal data levels of measurement

Values of ordinal variables have a meaningful order to them. For example, education level would be an ordinal variable (there is a definitive order to the categories, but we cannot make any other arithmetic assumptions beyond that). For instance, we cannot assume that the different in education level between undergraduate and high school is the same as the difference between graduate and undergraduate. We can use frequencies, percentages, and certain non-parametric* statistics with ordinal data.

ANOVA

used to measure THREE or more levels of IV

If a study is reliable, this means that? a. it was conducted by a reputable researcher who can be trusted b. it measures what it is intended to measure c. the findings can be generalized to others d. the DV causes the IV to change

c

Theresa is concerned that subjects in her experiment may respond in the way they believe they are intended to respond, rather than how they would normally. She is also afraid that her experiment may cause subjects psychological problems following their participation. Her first concern could be addressed by using a ___________ and her second concern could be addressed by using a _____________.

cover story; debriefing

Suppose you want to study perceptions of freedom of the press in the U.S. before and after the presidential election coming in 2005. Of the two types of designs, longitudinal or cross-sectional, which one would be most appropriate?

longitudinal

In my study I ask the same questions to the same people at several different points in time. The general type of survey I have used is _________ and the specific type is _________.

longitudinal; qualitative

Which type of survey would be appropriately classified as self administered:

mail

Choose the appropriate scale of measurement: Gender

nominal

The only difference between a static group comparison and a posttest only control group design is:

random assignment

Choose the appropriate scale of measurement: Height

ratio

Compared to open-ended questions, closed-ended questions:

require less effort from respondents and interviewer

Bad sampling frames, non-probability samples, and having some subgroup within the population fail to respond to your survey are all forms of:

sampling bias

Suppose I realize that many respondents have agreed with each statement on my survey. This tendency to consistently agreer is best described as:

Acquiscence response set

Ratio scale data levels of measurement

All arithmetic operations are possible on a ratio variable. An example of a ratio variable would be weight. We can accurately say that 20 pounds is twice as heavy as 10 pounds and exactly 0 pounds means that the object has no weight (meaningful zero-point). Other examples of ratio variables include gross sales of a company, expenditure of a company, income of a company, etc. A ratio variable can be used as a dependent variable for most parametric* statistical tests such as t-tests, F-tests, correlation, and regression.

One major different between true & quasi experimental designs is that with true experiments there is:

Randomization

A quasi experimental design is different from a true experiment because it may lack

Randomization & comparison group

The best way to control extraneous variables is to:

Randomly assign subjects to conditions (groups)

Suppose I was conducting a study to measure NJ residents' frequency of contact with foreigners. One result from my study is residents in New Brunswick contact foreigners three times more frequently than residents in Vineland. My level of measure (residents' frequency of contact with foreigners) is? A. nominal b. ordinal c. interval d. ratio

Ratio

Using college student volunteers limits the ______ of an experiment

external validity

Suppose I complain that Howard's experimental findings are only true in his experiments, and not in other contexts. My concern would be with _________ validity and probably indicates that Howard did his study in the _________.

external; laboratory

Choose the appropriate scale of measurement: ROM

interval

What type of question is this: How long have you been a fan of the Miami University football team?

open ended question

Choose the appropriate scale of measurement: Level of Assistance (eg. max assist +1)

ordinal

Choose the appropriate scale of measurement: MMT Grades

ordinal

Suppose I want to see what percent of Research I students think that the course will be easy before they take the class. Of the two types of designs, longitudinal or cross-sectional, which one would be most appropriate?

Cross-sectional

The researcher measures the ________. The researcher manipulates the ________. The researcher controls the __________.

DV, IV, extraneous variable

Is it desirable or undesirable to go to a Miami university football game?

Direct question

_________ surveys take the longest time to complete and __________ surveys generally are the least costly

Face to face; mail

Interval scale data levels of measurement

For interval variables, we can make arithmetic assumptions about the degree of difference between values. An example of an interval value would be temperature (we can assume that the difference between 70 & 80 deg. is the same as the difference between 80 & 90 deg.). Interval variables often do not have a meaningful zero-point. For example, a temperature of 0 deg. does not mean a complete absence of heat. An interval variable can be used to compute commonly used statistical measures such as mean, standard deviation, and the pearson correlation coefficient.

According to our discussion of a true experiment, which could NOT function as an independent variable?

Gender

Compared to experiments, surveys generally are:

Less effective in testing causal relationships

Which of the following is NOT an advantage to the face-to-face interview over the mail questionnaire?

Lower cost

If I tell you that there is a 5% chance that you will be selected in my sample, am I using a _________ sample?

Non-probability

What are the appropriate statistical tests for Nominal and Ordinal data?

Nonparametric tests (ex. chi square)

Epidemiologic studies do not involve manipulation of an independent variable, therefore it is this type of study

Observational experiment

What are the appropriate statistical tests for interval and ratio data?

Parametric tests (ex. t-test)

Suppose I conduct a survey which is only completed by the older adults in my sample? What type of error has occurred?

Sampling bias

Suppose I accidentally selected an extremely old person to be in my survey and only wanted ages 20-50 years old. What type of error has occurred?

Sampling error

Which statement about choosing a sample size is NOT appropriate:

Smaller samples are better if you are examining smaller relationships between variables

A test or instrument that measures what it is supposed to measure possesses: A. validity B. reliability C. consistency D. determinacy

validity