IDE 510, ISD 623, IDE 631

Question 1

What is Random Selection?

Answer 1

the process of selecting a representative sample of participants for a study from a larger population. All population has an equal chance of selection. This allows you to make generalizations about the larger population.

Reduces selection bias.
● Controls for extraneous variables because it increases the likelihood that confounding characteristics are equally dispersed
between the groups.
● Only used in experimental research because it is the key factor that allows one to make a strong claim of cause and effect from the
experiment. •

Question 2

Provide Examples of Weak Research Designs

Answer 2

One Group post test only (you don’t know if treatement has an effect becasue you don’t what it would be without the treatment)
One group pretest/post test design (better but does not control for extraneous var.)
Post test only with non-equivalent groups (paired t test or 1 way RM ANOVA) (Not randomly assigned. Does not account for diff. selection or attrition or additive /interaction effects

Question 3

Define Single Case Designs

Answer 3

ABAB Design- A design that is extended to include the reintrouction of the treatment condition
Multiple Baseline Design- the treatment condision is successfully administered to diff. participants or to the same participant in several settings after baseline behaviors have been recorded for diff. time periods
Changing Criterion Design- a participant’s beh. is gradually altered by changing the criterion for success of successive treatment periods

Question 4

Define Quantitative Variable

Answer 4

varies by degree or amount

E.g. annual income varies from 01- high income level

Question 5

Define Categorical Variable

Answer 5

Varies by type, group or kind

Question 6

Define constant variable

Answer 6

a single value or category of a variable; a variable is like a set of things (gender= male or female)

Question 7

Define Extraneous variable

Answer 7

a variable that may serve as a alternate explanation for change in the DV; competes with IV, should be controlled for to esbalish strong causal conclusion

Question 8

Define Confounding Variabile

Answer 8

an extraneous variable that gets entangled with the IV to cause a change in the DV. It confounds or confuses the causal relationship b/w IV and DV, means the researcher can’t tell if the target IV or confounding variable is the cuase for a change in the DV

Question 9

Control Variable

Answer 9

a variable that is controlled for in a study. It is known by the researcher and usually held constant across or between groups. (ex. controlling gender by only using males).

Question 10

Define Mediating variable

Answer 10

A variable in b/w the 2 variables of study in a causal chain. The IV influences the DV by way of the mediating variable. An ex. would be amount of studying (IV) leads to input and organization of knowledge in LTM (mediating) which affects test grades (DV)

Question 11

Independent Var.

Answer 11

the presumed causal variable. It is usually the var. that is manipulated. Manipulates the IV to test the hypothesis.

Question 12

Dependent Var.

Answer 12

the presumed effect var.; is usually the variable that is measured. AKA: outcome, criterion, response variable

Question 13

moderating Var.

Answer 13

a variable that changes the relationship b/w the IV and DV. Modification can include strengthening, weakening, or changing the direction.

Question 14

Define Nominal

Answer 14

A scale of measurement that uses symbols, such as words or numbers to label, classify, or identify people of objects (e.g. school type: 1- public; 2- private)

Question 15

Define ordinal

Answer 15

A rank order scale of measurement. This scale allows you to determine which person is higher or lower than another person on a variable of interest.

Question 16

Define interval.

Answer 16

A scale of measurement that has equal intervals of distance b/w adjacent numbers. It includes the rank order feature of ordinal scales and add. characteristics of equal distance. (temp scale., attitude).

Question 17

Define ratio.

Answer 17

A scale of measurement that has a true zero point. The scale includes properties of ordinal (rank order) and interval (equal distances between points) scales, plus has a true zero.

Question 18

Identify Differences between qualitative, quantitative and MM.

Answer 18

Quant:
Confirmatory or top down approach. Researcher tests hypothesis with data. Objective, agreed upon. Scientific. Justification by empirical confirmation of hypothesis. Regular and predictable.
MM:
Confirmary and exploratory. pluralism. appreciative of objective, subjective & intersubjective realities. pragmatic. What works for whom. Dynamic, complex, multiple influences.
Qualitative:
Exploratory researcher constructs hypothesis from knowledge. Subjective, mental, personal & Constructed. Relativism (individuals & group justification, varying standards); situational, social, contextual, personal unpredictable.

Question 19

Explain and differentiate 5 major quantitative and qualitative research methods

Answer 19

Phenomenology- In which the reseracher attempts to understand how 1 or more individual experience; unique perspective understanding rich data; subjectivity, bias, presentation, typical
Ethonography-describe cultural diff. of gap of people & to describe culture scenes; know of other cultures, understand human behavior, can evolve & discover new things, difficult to choose the sample, time consuming, relationships with residents & participants culture
Narrative-to inquire into people’s lives and told stories that can add to our understanding of people’s lives; flexible & deeper understaind of stubject; subjective bias
Case Study-describe 1 or more cases in depth; add alot of detail & can be used on rare cases; data can’t be generated; no scientific; small sample
Grounded Theory-generate a theory; thick description; & what actually happens; time consuming, researcher bias, not straight forward

Question 20

Define MM research.

Answer 20

Type of research in which a researcher mixes or combines qual. and quan. research philospophies/paradigms, methods, techniques into a single research study.
Eclectic, pragmatic & common sense approach. Mix in a way that works best for the research question. uses both deductive and inductive methods.

Question 21

Sampling Designs for MM Research

Answer 21

2 major criteria:
Time Orientation-
* Concurrent-data collected @ the same time
* sequential- data obtained in stages

Sample Relationship-

• Identical- same people participate in quan. & qual. phases of study.
• Parallel-separate quant. & Qual. samples drawn from population.
• Nonparallel-separate quant./qual. samples drawn from diff. populations
• multi-level- quan./qual. samples selected from different levels of a heirarchial population

Question 22

Discuss MM Validity.

Answer 22

Goal is to make justified quan./qual. claims.
External validity- ability to generalize to other populations. Not a priority in qual.
Research Bias is a threat.
Legitimation- addressing both qual./quan. threats to validity and trustworthiness

Question 23

Discuss Single Case Designs

Answer 23

Researcher attempts to demonstrate an experimental treatment effect using single participants.

Question 24

What is a non-experimental quantitative?

Answer 24

Research that lacks manipulation of the IV by the researcher. Researcher studies what naturally occurs or has occurred; how variables are related. Important in education.

Question 25

Define Quantitative Validity.

Answer 25

Internal validity- causal validity
External validity- generalization validity (to other populations)
Construct Validity- the extent to which a “construct” is accurately represented in a study

Question 26

Define Quantitative Validity

Answer 26

Descriptive- accuracy of accounts & facts
Interpretive- how well intepretations reflect meaning
Theoretical- how well the data collaborates conclusions
Evaluative- degree study was conducted with judgement bias
Generalizability- degree to which findings can be applied to population, situation, contexts, times.

Question 27

Define Independent variable

Answer 27

the program, policy, or process that is implemented in hopes of seeing change in knowledge, behavior attitud3, etc. the dependent variable.

Question 28

Define Null hypothesis:

Answer 28

typically the hypothesis of no difference between means or no relationship between variables in the population.

Question 29

Define Alternative hypothesis.

Answer 29

the logical opposite of the null hypothesis. It states that there is a relationship in the population

Question 30

What are threats to internal validity?

Answer 30

History- Any event other than the planned treatment that occurs 1, 2, measure of DV
Maturation- physical/mental changes over time that may affect performance on DV
Testing-any change in scores obatined on the 2nd admin. of a test as a result of having previously taken the test
Instrumentation- any change that occurs in the way DV is measured
Differential Selection-selecting partipants that have diff. characteristics for the various groups
Differential Attrition- loss of people that may not complete experient
Additive/interactive effects-2 or more basic threats to internal validity combined to produce a more complex bias

Question 31

Quasi-experimental methods are __________.

Answer 31

selection method in which participants are not randomly chosen or randomly assigned to the experimental and control groups

Question 32

What are the 3 types of hypothesis testing?

Answer 32

Pvalue approach, critical value approach and confidence interval approach

Question 33

Describe the p value approach to hypothesis testing.

Answer 33

1. State the Null and Alt.: H0: rho=0, H1: rho is not equal to 0.
2. set alpha level. Most researcher set the alpha at .05.
3. select the statis. test to be used. I will use the t test for the simple correlation coefficient.
4. Conduct the stat. test to be used and obtain pvalue. I ran the analys. and the observed corre. coeff. is .609 and p value is .001.
5. Compare the p value to alpha level and apply dec 1 or 2. Decision 1: if p is less than or equal to alpha then reject the null and tent. accept the alt. hypot. and conclude that the reserach is stat. sign.
   Rule 2: if p value is greater than the alpha, then fail to reject the null and conclude that the research finding is not st. sign.
6. Compute effect size, interpret findings, make judgement of practice sign. The corre. coeff. can be viewed as an effect size showing the magn. of relat.
   Cohen general rule sm=.1; medium=.3; lg=.5
   I consider finding to be prac. sign. because corre. is large and GPA is something students can congrol. If student increase GPA their start. salaries will be stat. significant.

Question 34

Describe the Critical Value approach to Hypothesis testing.

Answer 34

1. state Null and Alt.
2. Set Alpha
3. Select stat. test and test stat
4. Calculate the observed/obtained value of the test stat, and locate the critical value from the appropriate table in the back of the stat. book.
   Analysis and the observed value of the corre. coeff. was .609. the observed empirical value of t (calculated using the t test statistic) was 3.687. DF=N-2=25-2=23. The critical value for this test (df=23 and two tailed alpha =.05 was 2.069.
5. Compare the observed/obtained value of the test stat with the critival value from the sampling distribution and apply either decision 1 or 2.
   Rule 1; if the absolute value of the test is greater than the critical value then reject null and tentatively accept the alt. hypoth. and conclue research is stat. significant.
   Rule 2: if the abs. value of the test stat. is greater than the critical value, then fail to reject the null and conclus not stat. sign.
6. compute effect size, interpret findings

Question 35

P-value is_______________.

Answer 35

the probability of the observed value (or a more extreme value) of a stat assuming that the null hypothesis is true. (You can see that its a conditional probability, its a probability under the condition that the null is true).

Question 36

Random Selection_____________.

Answer 36

the process of selecting a representative sample of participants for a study from a larger population
● All population members have an equal chance of being selected.
● Allows you to make generalizations to the larger population (sample population)
● Very rare and difficult to do unless the population is small. o Also called “Random Selection”

Question 37

Types of Variables in QUantitative Research

Answer 37

Control Variable​: a variable that is “controlled for” in a study; It is known by the researcher and usually held constant across or between groups. Ex, controlling gender so only using males in the study.
Quantitative Variable​: a variable that varies by degree or amount
Categorical Variable​: a variable that varies by type, group, or kind
Constant:​ a single value or category of a variable; a variable is like a set of things (Gender= male or female) but a c

Question 38

What is Ratio?

Answer 38

the highest level of measurement where you still have rank order and equidistance, but HAVE true zero. An example is the Fahrenheit temperature scale, where the difference between 70 and 75 degrees is the same as the difference between 75 and 80 degrees. Note however that you cannot say that 80 degrees is twice as hot as 40 degrees because the zero point on an interval scale is arbitrary. (Can’t make ratio statements …twice as, half as much, etc…)

Question 39

What are principles of control in quantitative research?

Answer 39

Experimental Control​: the practice of controlling for extraneous variables through the design and conduct of the experiment; Best
approach is to use random assignment
Statistical Control​: used in non-experimental research; uses statistical techniques to remove the influence of extraneous variables.
● Partial Correlations- a special case of the GLM; a technique that partials out the influence of extraneous variables
● ANCOVA- another special case of GLM; a technique that shows the relationship between one Categorical IV, one
Quantitative IV, controlling for other quantitative extraneous variables.

Question 40

Name some types of control that can be used.

Answer 40

Matching​: used in non-experimental research; a researcher will often match participants in the treatment and control groups on a confounding variable; this controls the variable by holding constant and rules it out as a possible alternate explanation.

Building the variable into the design​: When this is done, the extraneous variable becomes another independent variable. This technique overcomes external validity problem of holding the extraneous variable constant because when you build the variable into your design, no group of participants is systematically excluded from your study…Factorial Design

Holding the variable constant​: The participants in each comparison group will have approximately the same type or amount of the extraneous variable

Counterbalancing:​ used in repeated measures designs; Used to offset sequence effects; The process of changing the order of treatment levels by group so that all groups receive all treatment levels, just in different orders; Can be partial or complete counterbalancing.

Question 41

Discuss MM research.

Answer 41

Mixed research methods is a type of research in which a researcher or team of researchers mixes or combines qualitative and quantitative research philosophies/paradigms, methodologies, methods, techniques, approaches, concepts, or language into a single research study or a set of related studies. The use of multiple disciplines is also integral to MMR.
Induction and deduction
Common sense approach
Mixed method research advises researchers to thoughtfully and strategically mix or combine qualitative and quantitative research methods, approaches, procedures, concepts, and other paradigm characteristics in a way that produces an overall design with multiple (divergent and convergent) and complementary strengths (broadly viewed) and nonoverlapping weaknesses.

Question 42

What is legitimation

Answer 42

In Mixed Methods research, validity requires addressing both quantitative and qualitative threats to validity and trustworthiness. These
must be addressed and resolved successfully.

Question 43

Discuss validity related to legitimation

Answer 43

● Inside-Outside Validity:​ the degree to which the researcher can present the “emic” or insider view as well as the “etic” or
objective outsider view.
● Paradigmatic-Philosophical Validity​: the degree to which the mixed method researcher articulates her “integrated”
philosophical and paradigmatic beliefs drawn from quantitative and qualitative methods.
● Commensurability Approximation Mixing Validity: ​ the degree to which the mixed method researcher can switch between the
lenses of a quantitative and qualitative researcher in order to produce a broader and integrated viewpoint.
● Weakness Minimization Validity​: the degree to which Quantitative and Qualitaive methods are chosen (strategically) to have
non-overlapping weaknesses; Select methods based on complementary strengths and weaknesses.
● Sequential Validity​: the degree to which a mixed researcher addresses any effects that may occur as the result of the order of
Quantitative and Qualitative phases in a study
● Conversion Validity:​ the degree to which quantitizing or qualitizing yields high-quality meta-inferences.
● Sample Integration Validity​: the degree to which a mixed researcher makes appropriate generalizations from mixed samples
● Socio-Political Validity​: the degree to which a mixed methods researcher incorporates interests, values, and viewpoints of
multiple stakeholders in a study.

Question 44

Discuss the paradigms of qualitative and quantitative research and their differences.

Answer 44

Pragmatism-

constructivism-

Question 45

Explain a 2 way interaction.

Answer 45

wo-Way interaction is present when the relationship between an IV and DV changes or varies at the levels of another independent variable. For example, the relationship between income and education can vary across the different levels of race in a society where institutionalized racism is commonplace. In this scenario, there could possibly be a significant relationship between income and education for those who fall in the dominant race category but there may be little or no significance for those who fall in the other racial categories.

Question 46

Explain 3 way interaction.

Answer 46

Three-way interaction is present when the interaction between two independent variables varies or changes at the levels of a third IV. For example, in a 2 x 2 x 2 where the first IV is Gender (Male, Female), the second is Experimenter (Nurturing, Exploitive) where subjects were told that the Experimenter was either exploitive or nurturing, and the third factor is Alcohol (At Risk for Codependency, Control) to denote subjects who had alcohol-dependent parents and those who did not. a three-way interaction is present when the interaction between the Alcohol factor and Gender varies or changes at the two levels of the Experimenter factor.

Question 47

Describe bivariate form of glm

Answer 47

1 quan dv, 1 cat iv
1way anova or t test if two categories

1 quan dv, 1 quant
Simple regression or pearson’s correlation coefficient