Brainscape's Knowledge GenomeTM

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


See full index

NUR3202C Research > Quanti - Inferential Statistics > Flashcards

Quanti - Inferential Statistics Flashcards

1
Q

What are inferential statistics?

A

make generalisations &
predictions on a population based on sample
data/results

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

What are the uses for inferential statistics?

A

Estimate parameters
eg. How many percent of Singaporeans are anxious about being infected
with COVID-19?

Test hypotheses
eg. Does Ginko nuts increase the memory of patients with Dementia?

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

Why do we need to estimate parameters?

A

To generalise sample characteristics to population
parameters that are often unknowable

Generalisation are still just estimations and we have to
account for inaccuracies and errors using confidence
interval (CI)
* CI: A range of values where the true mean lies
* Normally present mean, SD and CI

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

Describe the 6 steps for hypothesis testing.
(know what each step means)

A
  1. State null hypothesis (H0)
    * H0: μ = m0 (which means no effect, no relationship, no difference between true
    and observed mean)
  2. State alternative hypothesis(es) (H1)
    * H1: μ ≠ m0 (two-tailed: there is difference between true and observed mean)
    * H2: μ > m0 (upper-tailed: true mean > observed mean)
    * H3: μ < m0 (lower-tailed: true mean < observed mean)
  3. Set significance level (α)
    * Normally 0.05, which means that there is 5% chance that you will reject your H0when H0 is true
  4. Collect data
  5. Calculate statistics including p-value (probability of observing a sample statistic by
    chance, given that H0 is true)
    * E.g. if p-value=0.03, it means that 3 out of 100 times of your sample observation occurs by chance, given that H0 is
    true (meaning that it is unlikely that your observation occurred by chance)
    * If p-value=0.90, it means that 90 out of 100 times of your sample observation occurs by chance, given that H0 is true
    (meaning that it is very likely that your observation occurred by chance)
  6. Interpret results
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the 2 limitations of inferential statistics?

A

Can never be fully accurate because you are using
sample data to estimate/infer that of a population

Interpretation of data is subjected to the researchers’
reasoning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the characteristics of parametric tests?

A
  • Follows a normal distribution
  • Central tendency measure - assessed group mean
  • Continuous variables
  • Higher statistical power
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are non-parametric tests?

A

Normally for results in ordinal data (e.g. Likert scale)
but can also be used for continuous data (i.e. ratio or
interval e.g. age) that are not confirmed to assume a
normal distribution.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the characteristics of non-parametric tests?

A
  • No need to follow normal distribution (mostly based on rank order or how common data is)
  • Central tendency measure - assess group median
  • Can be used for both continuous and discrete variables
  • Lower statistical power
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the common assumptions for parametric tests?

A
  • DV is continuous (interval/ratio)
  • DV follows a normal distribution
  • Homogeneity of variance between groups
  • Comparison groups are independent (subjects in both groups cannot influence each other)
  • Preferably no significant outliers
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How do we check for normality?

A
  1. Visualization
    - Q-Q plot
    - Histogram
  2. Statistical hypothesis testing
    - Shapiro-Wilk test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the empirical rule?

A

68-95-99.7 rule

3 percentage values for which values lie within 1,2 and 3 SD

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What can you do is your sample distribution violates normality assumptions?

A
  • Use non-parametric tests
  • Transform your data (eg. log, square etc)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the parametric and non-parametric test for the following?

One group mean/median

A

Parametric:
One sample t-test/z-test

Non-parametric:
Wilcoxon Signed Rank test

Eg.
Is the mean age of menopause in Singaporean women 49 years old?

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

What is the parametric and non-parametric test for the following?

Two means/medians from the same person/group

A

Parametric:
Paired sample t-test

Non-parametric:
Wilcoxon Signed Rank test

Eg.
Would there be a significant change in blood glucose level of the intervention group between baseline and one-year follow-up?

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

What is the parametric and non-parametric test for the following?

Two independent group means/medians

A

Parametric:
Indepenent samples t-test

Non-parametric:
Mann-Whitney U test

Eg.
Are there signifiant differences in anxiety levels between undergraduate nursing in year 1 compared to those in year 4?

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

What is the parametric and non-parametric test for the following?

≥ Three independent group means/medians

A

Parametric:
One-way Analysis of variance (ANOVA)

Non-parametric:
Kruskal-Wallis test

Eg.
Are there significant differences in BMI among adults who underwent bariatric surgery, exercise intervention, diet intervention or no intervention (control group)?

17
Q

What is the parametric and non-parametric test for the following?

≥ Three independent group means/medians of two independent variables (categorical)

A

Parametric:
Two-way ANOVA (eg. repeated measure ANOVA)

Non-parametric:
Friedman’s test

Eg.
Are there significant differences in blood pressure between those wit high, moderate and low level of physical activity at 3 month, 6 month and 9 month follow-ups?

18
Q

What is a one sample t-test?

A

Compares sample mean to pre-defined/a priori mean.

eg. Is the mean life expectancy of healthcare professionals
the same as the population mean of 83.2 years?

19
Q

What is an independent sample t-test?

A

Compares sample means between two groups.

20
Q

What is a two-way ANOVA?

A

Commonly used to find out whether there are changes in
DV due to an interaction effect between two IV (e.g. over
≥2 time points under ≥2 conditions [e.g. treatment &
control groups]

E.g. Are there significant differences in blood pressure
between the intervention and control group at 3-month,
6-month and 9-month follow-up?

21
Q

What is the Wilcoxon signed-rank test?

A

To find out if the median is equals to a specific value e.g.
* H0: The median pain score = 2
* H1: The median difference ≠ 2

Assumptions:
* Differences are independent and randomly obtained
* Sample difference is symmetrically distributed (does not mean normally distributed-around same
number of values near median)

Steps:
* Calculate differences between sample data and specific value (or two different samples)
* Rank differences regardless of sign (+/-)
* Assign corresponding signs (W+ and W
-) & calculate total sum of W+ and W-
* Compare the smaller test statistic (W) with critical value (s). If < critical value, reject H0

22
Q

What is the Mann-Whitney U test?

A

To find out if there are differences in distribution/medians
between two groups:
* H0: no differences between the two groups’ medians
* H1: there is difference between the two groups’ medians

Assumptions:
1. DV is of ordinal or continuous level of data
2. IV has two independent groups (e.g. male/female, yes/no)
3. Groups are independent of each other
4. IV groups follow similar distributions

23
Q

What is the Kruskal-Wallis test?

A

Same as Mann-Whitney U test except that it compares
three or more independent groups
* H0: no difference among the groups
* H1: there is difference in at least one group in comparison with the
others

24
Q

What is the Friedman’s test?

A

To find out if there are differences in repeated measures
e.g. in treatment effect across time points
* H0: There is no significant difference among the groups (time-points)
* vs H1: There is significant difference among the groups (time-points)

25
Q

Why do we measure linear association?

A

Produces a coefficient used to quantify the strength and
direction of a relationship/association between two or
more variables

Value of coefficient ranges from -1 to +1.

26
Q

What is the Chi-square test of independence?

A

Measures significance of an association between two
categorical variables

E.g. Among patients with Hodgkin’s disease, is there a
significantly disproportionate number of females than
males, indicating a significant association between
haematological cancer and sex?

27
Q

What is linear regression?

A

An estimation of the association between a continuous
DV and ≥1 IV
* Assumptions:
* Linear relationship
* Independence
* Homoscedasticity
* Normality

28
Q

What is the formula for simple linear regression?

A

𝑦 = a + b𝑥 + 𝑒

E.g. Lung function
* Forced expiratory volume=
0.0125 + 0.364(age) + 0.0135

29
Q

What is the formula for multiple linear regression?

A

𝑦 = a + b1𝑥1 + b2𝑥2+b3𝑥3+…+ e

E.g. Lung function (Ma et al., 2013)
* Forced expiration volume= -2.63 + 0.108(age) + 0.011(weight) +
0.021(height) + 0.738

30
Q

What is logistic regression?

A

An estimation of the association between a binary DV and ≥1 IV

31
Q

What is the purpose of correlation vs regression?

A

Correlation: Describe, infer
Regression: Predict

32
Q

What is the parametric test for correlation?

A

Pearson’s correlation coefficient

33
Q

What is the non-parametric test for correlation?

A

Spearman rank-order correlation coefficien

34
Q

What are the ranges for effect size strength (Cohen’s standard) for correlation & regression?

A

Very weak: r < 0.3
Weak: 0.3 < r < 0.5
Moderate: 0.5 < r < 0.7
Strong r > 0.7

NUR3202C Research (12 decks)

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