1/80
Flashcards based on Quantitative Data Analysis I lecture notes.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
What are the four types of data based on measurement properties?
Nominal, ordinal, interval, and ratio.
What is the key difference between parametric and non-parametric tests?
Parametric tests assume data is normally distributed, while non-parametric tests do not.
What type of test is used to determine if data is normally distributed?
Normality test (e.g., Shapiro-Wilk test).
Name three statistical tests used for testing differences between two groups.
Student’s t-test, Mann-Whitney U-test, and Chi-squared test.
What is an explanatory variable?
A variable that is expected to be a cause.
What is a response variable?
A variable that is expected to be an effect.
Give an example of a nominal variable.
Hair color (e.g., red, blonde, brown).
Give an example of an ordinal variable.
Likert scales (e.g., strongly disagree to strongly agree).
Give an example of an interval variable.
Temperature (Celsius or Fahrenheit).
Give an example of a ratio variable.
Weight.
What type of data is categorized and ranked, but not evenly spaced?
Ordinal data.
What type of data is categorized, ranked, evenly spaced, and has a natural zero?
Ratio data.
What type of data is categorical and not ranked?
Nominal data.
What is a dichotomous variable?
A variable with only two categories (e.g., yes/no).
Which type of statistical test is more powerful?
Parametric tests are generally more powerful than non-parametric tests.
What does a normality test assess?
Whether the distribution of the data is normal or skewed.
Name a graphical method for assessing normality.
Histogram plot.
Name a numerical test for assessing normality.
Shapiro-Wilk test or Kolmogorov-Smirnov test.
What does a significant p-value in a Shapiro-Wilk test indicate?
Deviation from normality.
Which test should you perform if p-value is higher than 0.05 in Shapiro-Wilk test?
Parametric test
What is the null hypothesis (H0) in the Shapiro-Wilk test?
The data distribution is not different from a normal distribution.
What is the alternative hypothesis (H1) in the Shapiro-Wilk test?
The data distribution is different from a normal distribution.
What is checked when testing for differences between two groups?
Independent variable
What are the three assumptions of Student's t-test?
Random samples, normally distributed data, and homogeneity of variance.
What does Levene's test check?
Homogeneity of variance (homoscedasticity).
In Student's t-test, a larger T value indicates what?
A stronger trend (larger difference between means, smaller standard deviation).
What probability indicates statistical significance in Student's T-test?
P=0.026 (an example)
If the dependent variable deviates from normality, can a T-test preformed?
No
Which test should you perform if the dependent variable is numerical, but is not normally distributed?
Mann Whitney U-test
On what data the Mann Whitney U-test performs the test?
Data is ranked from low to high
What is reported when performing Mann Whitney U-test?
Differences between groups reported as medians and not means
With what kind of data the Chi-squared test should preform the test?
Nominal data/categorical variable
What are the expected values in Chi-squared test?
Calculated (theoretical) frequencies that would be expected if the null hypotheses were true – no difference between groups.
What are the observed values in Chi-squared test?
Actual values from your data collection (counts or frequencies)
The higher perception of taller than average GSS students are examined by which test?
Chi-squared test
What is the formula for the expected values in Chi-squared test?
(column total*row total)/grand total
What does the degrees of freedom in X2 indicates?
Number of observed columns minus one, multiplied by the number of observed rows minus one
Is the consumption of friut associated with frequency of physical activity?
Analyzed by Chi-squared test
What does 'location parameter' indicates in Mann Whitney U-test?
Difference between groups is significant
What is a continuous variable?
Numerical data.
What is another term for independent variable?
Explanatory/exposure variable.
What is another term for dependent variable?
Response/outcome variable.
Is pain rating in a 1-10 scale qualitative or quantitative?
Quantitative.
Is hair color considered numerical data?
No, it's non-numerical.
What does 'increasing degree of information' indicate?
More arithmetical operations.
What operation can be applied on nominal data?
Frequency or percentage.
If data is not normally distributed, what type of test is appropriate?
Non-parametric test.
What descriptive statistics are used for central tendency?
Mean, median, and mode.
What descriptive statistics are used for dispersion?
Range, variance, and standard deviation.
Define skewness.
A measure of the asymmetry of the probability distribution of a real-valued random variable about its mean.
What are the two approaches to test for normality?
Graphical and numerical.
Pros of graphical assessment of normality.
Easy and quick.
Cons of graphical assessment of normality.
Experience needed and not always objective.
Cons of numerical assessment of normality.
Unsensitive at small sample sizes and overly sensitive at large sample sizes.
What should you do when still in doubt of data normality?
Run both parametric and non-parametric versions of same statistical test.
What is the H0 of Shapiro-Wilk?
The data distribution is not different from a normal distribution.
What does a significant p-value in Shapiro-Wilk test indicate?
Deviation from normality.
When is a parametric test not applicable?
When data deviates from normality.
What are the three types of questions used as examples as the lecture?
Willingness to install solar panels, willingness to use CO2 neutral forms of transportation, the effect of wind stress on plant diversity.
What is the key component of 'Differences between two groups'?
Independent variable = two groups.
If the p-value > 0.05 in Levene's test, what does that indicate?
The variances are equal.
What is ranked from low to high in the Mann Whitney U-test?
Data.
Research question: Is the perception of corruption in Africa the same as in Asia?
Dependent variable: Perception of corruption, Independent variable: Country.
In Chi square test What does comparing observed values with expected values test?
Test relationship between two categorical variables
If there is no difference in preference among the groups what can be said about expected values?
Expected values are calculated frequencies that would be expected if there were no difference between groups
What does GSS stands for?
I am sorry, this was never defined in the text.
What indicates that students taller than average are more likely to indicate to be taller than average?
H0: Students taller than average are as likely to indicate to be taller than average than students shorter than average.
What is the difference between Yes (observed) and Yes (expected)?
Yes(observed) are just the amount of people who said yes and were shorter or taller, Yes(expected) accounts that it is by chance.
What does the formula χ2 = ∑(f1-e1) 2/e1 indicates?
That students taller than average perceived themselves as taller than average more frequently than students that are shorter than average.
What are the options to check in JASP for Mann-Whitney test?
✓ Mann-Whitney ✓ Location parameter ✓ Effect size
To what is the frequency of fruit consumption associated?
Frequency of physical activity
The summary part consists of?
Part 1: (more) Fundamental terminology, Part 2: Testing differences between two groups
What statistical test is used with nominal data and two groups?
Chi-square test.
What statistical test is used with ordinal data and two groups?
Mann-Whitney U test.
You have two related sets of data and want to examine the association between them. What test could you use?
Spearman’s R or Pearson’s R
Which test requires normal distribution of scale?
T-test.
Is Spearman’s R parametric of non-parametric test?
Non-parametric
Is Shapiro-Wilk test graphical or numerical?
Numerical.
Is normality test objective or subjective?
Numerical normality test is objective.
Is statistical test run before or after plotting histogram to see the data distribution?
After.
If results differ between parametric and non-parametric tests, what actions should be taken?
Be very careful in your interpretation of the results.