11. Practice Evaluation & Utilization of Research

Question 1

What is a research design?

Answer 1

Research design is viewed as a logical plan to help increase our knowledge in a particular area.

Question 2

Why design research?

Answer 2

There are two major purposes for research design:
 To provide answers to research questions
 To control variance, insuring that the results obtained are as closely related to what is being studied.

Question 3

What is a variable?

Answer 3

A variable is any phenomena or characteristic that is free to vary with at least 2 conditions or levels. A constant is restricted to a single state.

Question 4

How should variance be controlled?

Answer 4

This involves the confidence level that can be placed in accurately assessing the degree of cause and effect you have as a result of the research endeavor. This is often referred to as the Maximincon principle.

Maximize the variance of variables of the substantive research hypothesis (i.e. you want the experimental conditions to be as different as possible); to control extraneous variables (i.e. these are independent variables that are not necessarily related to your study (e.g. sex, race); and minimize error variance (i.e. This is the fluctuation that may occur in your results due to random events).

Question 5

What is an example of a variable?

Answer 5

Demographic variables are used to define your sample

Question 6

What are the two types of variables?

Answer 6

Two types of important variables are the independent and dependent.
 Independent variables are the presumed cause.
 Dependent variables are the presumed effect and vary as related to the independent variable. Usually the treatment being tested is the independent variable.

Question 7

A research problem should be describes using a hypothesis. What is a hypothesis?

Answer 7

A hypothesis is a proposition that is stated in testable form. It predicts a particular relationship between two or more variables. If we think a relationship exists, we must first generate a hypothesis and proceed to test it. There can be more than one hypothesis. Generally, a hypothesis follows an “if” then “this” will occur format. If such-and-such occurs then so-and-so results.

The null hypothesis (AKA statistical hypothesis) is often referred to as the hypothesis of “no difference.” This is hypothesis is generally used in statistical analysis to be disproved. The research (AKA alternate hypothesis) or the working hypothesis is the hypothesis you will be testing. The research hypothesis is what you want to support.

Question 8

What is a Random Sample?

Answer 8

Random sample: Briefly stated, it is a planned process that utilizes probability theory to ensure that the sample will represent the population. In the random sample each subject in the population has an equal chance of being selected.

Question 9

What is Generalizability?

Answer 9

Generalizability: this is the true goal of all research. It is where you take what you know about a small group or sample of a population and apply it to explain the general population.

Question 10

What is Generalizability?

Answer 10

Generalizability: this is the true goal of all research. It is where you take what you know about a small group or sample of a population and apply it to explain the general population.

Question 11

What are Inferential statistics?

Answer 11

These robust powerful statistics help the researcher make “inferences” or assumptions about a population. These have strong rules & strong guidelines.

Question 12

What are some examples of inferential statistics?

Answer 12

Parametric:
• Analysis of variance aka ANOVA aka F test (compares means of more than two groups)
• T test (compares means of two groups
• Pearson’s Pho or Pearson’s R (compares the association or correlation between two groups

Question 13

What are Descriptive statistics?

Answer 13

Descriptive statistics : examples are nonparametric tests (no strong rules or guidelines, less confidence) include:
 Chi-squared test (most common type, compares the observed value with the expected)
 Spearman Rho (a non-parametric correlation)

Question 14

What is correlation?

Answer 14

Correlation: Means there is an association between two variables:
 Positive association: as one goes up, the other goes up; as one goes down, the other goes down.
 Negative or inverse association: as one goes up, the other goes down; as one goes down, other goes up.

Question 15

How is association (of correlation) measured?

Answer 15

Correlation measures the mathematical relationships between two variables.
• When you have a perfect correlation it =1 (that means they move in the same trend direction at the same time. The closer the number to 1 the better the correlation. A .9 is a very good correlation. A .5 is pretty poor

Question 16

What is statistical significance (confidence)?

Answer 16

Statistical Significance:
 This is your “educated quotient” the research standard is usually set at .05
 A probability of .05 means that 5 times out of 100 you will be wrong and 95 times out of 100 you will guess correctly.
 A probability of .02 means two times out of 100, you will be wrong and 98 times you will be considered to have guessed correctly.

Question 17

What are tome common statistical distributions of values (curves)?

Answer 17

Bell-shaped curve: is a symmetrical distribution (values most likely to occur are close to the mean/average).

Kurtosis curve: it is a skewed distribution (values are more likely to occur along the spectrum of possibilities - not likely to be close to the average)

Question 18

How are practical experiments designed?

Answer 18

An experimental design is ordinarily not possible in a practice setting, as there are ethical, policy and practice constraints against denying services to people in need. Since a control group requires an equivalent population that did not receive the intervention, an experimental design is usually impractical for most practice setting. Quasi-experimental designs allow variations in the intervention, in time, and in the population served, and are more generally used in practice research.

Question 19

What is a type 2 error?

Answer 19

Type 2 refers to a category of statistical error.