Data analysis and interpretation of findings

Question 1

identify the levels of data measurement in a research study

these are descriptive

Answer 1

1) nominal
2) ordinal
3) interval/ratio

Question 2

Nominal level of measurement

Answer 2

> lowest level
“dummy variables used to quantify
male/female, yes/no

one is not higher than the other. they are worth the same

> cannot do a T-test or linear test on this

Question 3

Ordinal measurement

Answer 3

> sorting of objects on the bases of their standing on an attribute relative to each other. 
>A higher score is better (or worse). 
> class ranking, likert scale responses
> strongly agree--> strongly disagree
> rate your level of satisfaction

> if use a number 1–>5, then it can be argued that it is ratio/interval

Question 4

interval/ratio measurement

Answer 4

> determines both the ranking order of objects on an attribute and the distance btwn them
scores on intelligence test, temp, BP, height, length
allows more manipulation of data and calculation of means

> ratio is the highest level of measurement, but usually only achieved in physical sciences
height, weight, pulse, BP

Question 5

Distinguish descriptive and inferential statistics

Answer 5

descriptive: are obtained through descriptive techniques that reduce data to manageable proportions by summarizing and organizing them.
> used to describe characteristics of the sample and describe values obtained

> can arrange data “visually”

> measures of tendency: mode, median, mean, scatter plots, %, graphs

Inferential: allow researchers to to estimate how reliably they can make predictions and generalize findings on the basis of the data.
>combine math processes and logic to test hypothesis about a population with the help of sample data.

> stats that permit inferences on whether relationships observed in a sample are likely to occur in a larger population.

Question 6

Distinguish the characteristics and uses of parametric and nonparametric statistics

Answer 6

Parametric: assumes normal distribution (ex. random sample)
> preferred by researchers, more powerful**
> can formulate simple sample statistics, such as mean and SD
> interval or ratio level of data
> ex. T-test, ANOVA, Regression, Pearsons product moment correlation

Nonparametric: does not assume normal distribution within sample (ex. not random sample)
>distribution-free tests
> nominal or ordinal level of data
> ****less powerful
> ex. Chi-square test,

Question 7

Discuss the factors to consider when interpreting research findings

Answer 7

1. are the results of each hypothesis presented?
2. is the info regarding the results concisely and sequentially presented?
3. are the tests that were used to analyze the data presented?
4. are the results presented objectively?
5. if tables or figures are used, do they meet the follolwing standards? proper text and headings…
6. are the results interpreted in light of the hypothesis and theoretical framework and all of the other steps that preceded the results?
7. talk about weaknesses and strengths
8. discuss clinical relevance?
9. generalizations made?
10. recommendations?
11. what was the studies strength of evidence?
12. what was the level of significance? (0.05 alpha)
    > probability of making an error