Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Research 3900 > Unit 3: Ch. 12 > Flashcards

Unit 3: Ch. 12 Flashcards

1
Q

descriptive statistics

A

used to describe and synthesize data; typically used to describe demographic data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

inferential statistics

A

used to make inferences about the population based on the sample

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

frequency distributions

A

systematic arrangement of numeric values on a variable from the lowest or the highest with a count or percentage of the number of times each value occurred
-ex: 57%, 5%, 28%, n=27

Frequency distributions can be described in terms of:

  • shape
  • central tendency
  • variability

Can be presented in a table (Ns and percentages) or graphically

Standard deviation: abbreviated as “SD” in articles

  • average deviation of scores in a distribution
  • small SD indicates consistency among respondents; large SD indicates wide variation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

standard deviation

A

abbreviated as “SD” in articles

  • average deviation of scores in a distribution
  • small SD indicates consistency among respondents; large SD indicates wide variation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

frequency distributions: shape

A

symmetric or asymmetric

asymmetric = “skewed”

skew is a rough indicator of the normality of the distributed data; reflects the degree w/ which scores on a variable fall at one end of the other on the scale of a variable

  • normality: under a normal curve, it’s a normal distribution (“bell shaped curve”)
  • skew test: ex test scores
  • if the majority of scores fall at the high end of the scale, the distribution is negatively skewed
  • if the majority of the scores fall at the low end of the scale, the distribution is positively skewed
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

frequency distributions: central tendency

A

an index of the typicalness of a set of scores that comes from the center of the distribution
-ex: normal curve (“middle 2/3 of the normal curve”)

3 measures:

  • Mode: most frequently occurring score in distribution
  • Median: point in the distribution in the middle
  • Mean: average; in articles you’ll see “M” for mean or the word “average”; may see “x-bar” (an “x” w/ a line over it)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

frequency distributions: variability

A

degree to which scores in a distribution are spread out or dispersed

  • homogeneity: when there is little variability in the group of scores you have (“everybody is roughly getting in the same few scores”)
  • heterogeneity: a lot of variability (ex: exam scores)
  • Range: highest value to the lowest value
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

bivariate descriptive statistics include? (2)

A
  1. cross tabs (aka contingency table)

2. correlation coefficients

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

bivariate descriptive statistics: Cross Tabs (Contingency Table or 2-way ANOVA)

A

evaluates whether a statistical relationship exists between 2 variables, each of which has different levels

  • ex: types of maltreatment (maltreatment is a “construct” with 5 levels); demographic variables (whether or not someone is a HS graduate, whether or not they’re employed)
  • researcher ends up with the probability of a thing happening
  • -> ex: “adolescent mothers who have been sexually abused were 2.5 times more likely to be high school graduates and employed”
  • produces an odds ratio
  • -> odds ratio: the odds of one thing happening versus the odds of another thing happening
  • —> ex: high school graduation: either you graduate or you don’t, odds ratio gives probability of that happening
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

bivariate descriptive studies: Correlation Coefficients

A

describe the intensity and direction of a relationship between 2 variables

intensity: represented by the absolute value of a number

direction of the relationship is indicated by either a positive (+) or a negative (-) sign

the higher the absolute value of a number the stronger the relationship

  • ex: number of 0.25 is much weaker than relationship with 0.80
  • when looking at the strength of the relationship you’re not looking at the +/- sign –> you’re just looking at the number itself (the absolute value)

Correlation uses the “r” statistic (“r=”)

Negative relationship: inverse relationship; when one variable goes up the other variable goes down

Positive relationship: when one variable goes up the other variable goes up

R = -0.45 and -0.80 –> -0.80 is stronger relationship

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

describing risk

A

clinical decision making for EBP may involve the calculation of risk indexes, so that decisions can be made about relative risks for alternative tx or exposures

Some frequently used indexes:

  • absolute risk
  • absolute risk reduction (ARR)
  • odds ratio (OR)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

absolute risk

A

the proportion of people who experienced an undesirable or desirable outcome
-ex: proportion of smokers who get lung cancer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

absolute risk reduction (ARR)

A

the difference between the absolute for exposure to an outcome and the absolute risk for no exposure to an outcome (“exposure minus no exposure”)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

odds ratio (OR)

A

proportion of people w/ an adverse outcome relative to the people w/o the adverse outcome (“probability”)

ex: OR = 1.75

may be seen in conjuncture with confidence intervals

look at p. 225 of textbook for examples of OR

needs to be statistically significant to mean that the result didn’t happen by accident

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

inferential statistics

A

used to make objective decisions about population parameters using the sample data
-using a sample to make decisions about the whole population

based on laws of probability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Point estimation

A

a single, descriptive statistic that estimates the population value (e.g. a mean, percentage, or OR)
-basically saying you have one number you’re dealing with

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

SEM (standard error of the mean)

A

used to determine the SD of the sampling distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Interval Estimation

A

a range of values within which a population value probability lies
-involves computing a confidence interval (CI)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

confidence interval (CI)

A

indicates an upper and lower range; looking at the probability of a particular score you get being within that range

ex: 95% CI of 40-60 for a sample mean of 45 indicates there’s a 95% probability that any one score will be between that range

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

parametric statistics

A

involves the estimation of a parameter; assumes variables are normally distributed in the population (“not skewed”); measurements are on interval/ratio scale

use for big samples

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

nonparametric statistics

A

does not involve estimation of a parameter; measurements typically on nominal or ordinal scale; doesn’t assume normal distribution in the population

use for small samples

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

type 1 error

A

rejecting a null hypothesis when it’s true

  • getting a false positive
  • saying someone has a condition/dz when they don’t have it (ex: false positive pregnancy test)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

type 2 error

A

accepting the null hypothesis when it’s false

  • getting a false negative
  • saying that someone doesn’t have a condition/dz when they really do
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

level of statistical significance

A

significance does not mean important or meaningful
-means it didn’t happen by accident

level of significance is set by the researcher

p < 0.05 - means that the finding was statistically significant (“didn’t happen by accident/chance”)
-“95% chance that the findings are real”

p < 0.01 - means that the finding was statistically significant (“didn’t happen by accident/chance”)
-“99% chance that the findings are real”

25
Q

Bivariate statistical tests include? (4)

A
  1. t-tests
  2. Analysis of variance (ANOVA)
  3. Correlation coefficients
  4. Chi-squared test
26
Q

t-tests

A

uses the “t” statistic, tests the difference between the means of 2 groups

  • asking if the group means are different from each other
  • the mean can be from 2 independent groups (such as men vs. women) or from 2 dependent groups or paired groups (such as preop vs. postop)
27
Q

analysis of variance (ANOVA)

A

tests the difference between more than 2 means; “a whole family of statistics”

uses the “f” statistic

one way ANOVA: 3 different groups researchers are testing the means on

two way ANOVA (multi-factor ANOVA): contingency table or cross tabs

repeated measures ANOVA (RM-ANOVA): may be used on longitudinal study

28
Q

correlation coefficients

A

correlations test the relationship between variables; “Pearson’s Product Moment;” Pearson’s uses the “r” statistic

Parametric Version: “Pearson’s;” large sample

Nonparametric Version: “Spearman’s Rho;” small sample; may be used to confirm a correlation

29
Q

Chi-squared test

A

tests the proportions of categories within a contingency table; looking at frequencies; comparing observed frequencies
-measured on nominal or ordinal scale

chi-squared: x^2

computed by summing differences between the observed frequencies in each cell and the expected frequencies - those that would be expected if there were no relationship between the variables

30
Q

multivariate statistical analysis

A

statistical procedures for analyzing relationships among three or more variables

commonly used procedures in nursing research:

  • multiple regression
  • analysis of covariance (ANCOVA)
  • multivariate analysis of variance (MANOVA): procedures for analyzing relationships among 3 or more variables
  • logistic regression

*be familiar with the names of these rather than what they do

31
Q

effect size

A

an important concept in power analysis

  • asks “how big was the intervention or treatment on the outcome?”
  • “if you have an intervention you’re going to have an effect size of some kind”
  • power analysis
  • known relationship: use a 1 tailed test
  • unknown relationship: use a 2 tailed test

effect size indexes summarize the magnitude of the effect of the independent variable on the dependent variable

seen mostly in SYSTEMATIC REVIEWS

in a comparison of 2 groups (i.e., in a t-test situation), the effect size index is “d”

by convention:

  • d ≤ 0.20 = small effect
  • d = 0.50 = moderate effect
  • d ≥ 0.80 = large effect
32
Q

power analysis

A

estimate of the sample size needed to detect an effect from the IV on the DV (if there is one)

  • tells you the right sized sample so if there’s an effect from the tx you’ll pick it up
  • “49 participants achieved a power of 80 with a moderate effect size of…”
33
Q

When indexes such as averages and percentages are calculated with data from a population, they are ____.

A

parameters

34
Q

A descriptive index from a sample is a ____.

A

statistic

35
Q

unimodal distribution

A

one peak

36
Q

multimodal distribution

A

2 or more peaks

37
Q

bimodal distribution

A

2 peaks

38
Q

univariate descriptive statistics

A

one-variable

ex: mean or SD describe one variable at a time

39
Q

bivariate descriptive statistics

A

describe relationships between 2 variables

40
Q

correlation matrix

A

variables are displayed in both rows and columns

41
Q

Pearson’s r

A

aka product-moment correlation coefficient

computed with interval or ratio measures

most commonly used correlation index

there are no fixed guidelines on what should be interpreted as strong or weak relationships, b/c it depends on the variables

42
Q

parameter estimation

A

used to estimate a population’s parameter

-ex: a mean, proportion, or a mean difference between 2 groups (e.g., men vs. women)

43
Q

sampling distribution of the mean

A

it’s theoretical: in practice, no one actually draws consecutive samples from a population and plots their means

44
Q

statistical hypothesis testing

A

uses objective criteria for deciding whether research hypotheses should be accepted as true or rejected as false

45
Q

The rejection of the null hypothesis is what researchers seek to accomplish through ____ ____.

A

statistical tests

46
Q

null hypothesis

A

states there is no relationship between the independent and dependent variables

47
Q

power

A

the ability of a statistical test to detect true relationships, and is the complement of beta (that is, power equals 1 minus beta).

48
Q

In hypothesis testing, researchers use study data to compute a ____ ____.

A

test statistic

49
Q

statistically significant

A

results are not likely to have been due to chance

50
Q

nonsignificant results (NS)

A

any observed difference or relationship could have been the result of a chance fluctuation

51
Q

p level

A

probability

52
Q

hypothesis testing process (5)

A
  1. selecting a test statistic
  2. specifying the level of significance
  3. computing a test statistic
  4. determining degrees of freedom (df)
    - refers to the number of observations free to vary about a parameter
  5. comparing the test statistic to a theoretical value
53
Q

post hoc tests

A

aka multiple comparison procedures

Statistical analyses known as post hoc tests are used to isolate the differences between group means that are responsible for rejecting the overall ANOVA null hypothesis

54
Q

Researchers can improve their prediction of an outcome by performing a ____ ____ in which multiple independent variables are included in the analysis.

A

multiple regression

55
Q

multiple correlation coefficient

A

the coefficient in multiple regression

symbolized as “R”

R varies from .00 to 1.00 (NO negative values), showing the strength of the relationship between several independent variables and an outcome, but not direction.

researchers usually report multiple correlation results in terms of R^2 rather than R

56
Q

analysis of covariance (ANCOVA)

A

combines features of ANOVA and multiple regression; used to control confounding variables statistically
-“equalize” groups being compared

57
Q

logistic regression

A

analyzes the relationships between multiple independent variables and a nominal-level outcome (e.g., complaint vs. noncomplaint)

58
Q

T/F: Hypotheses should be described as being supported or not supported, accepted or rejected.

A

true

Research 3900 (15 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions