LECTURE 8 Flashcards
What is correlational analysis?
A method to examine associations between two variables.
What is a scatterplot?
A graph showing associations between two variables with each dot representing a data point.
What is the direction of association?
Indicates whether the variables are positively or negatively related.
What is a positive association?
Larger values of one variable are associated with larger values of the other.
What is a negative association?
Larger values of one variable are associated with smaller values of the other.
What is the strength of an association?
Reflects how much knowing one variable helps predict the other.
What does no association mean?
Knowing one variable gives no information about the other
What is the shape of an association?
The pattern of the relationship, such as linear, U-shaped, or inverted U-shaped.
What are correlation coefficients?
Descriptive statistics that describe the strength and direction of an association.
What is the Pearson correlation coefficient used for?
Linear associations between normally distributed interval or ratio variables.
What is the Spearman correlation coefficient used for?
Linear associations between ordinal data or interval/ratio data that is not normally distributed.
What is the range of correlation coefficients?
From -1 (perfect negative) to +1 (perfect positive), with 0 indicating no association.
What does the numerator in Pearson correlation measure?
Covariance, which indicates the direction of the association.
What does the denominator in Pearson correlation measure?
Standardization, allowing interpretation independent of the scale.
Do linear transformations affect Pearson correlation?
No, changes like converting cm to m do not impact the coefficient.
What is the Spearman correlation based on?
Ranks of data instead of raw values.
What is an example of Spearman correlation?
Investigating the relationship between fear of missing out (FOMO) and pub visits.
How is Spearman correlation calculated?
By ranking the data and computing the Pearson correlation on the ranks.
What is a two-sided test in correlation analysis?
Tests whether the correlation in the population is different from zero.
What is a one-sided test in correlation analysis?
Tests if the correlation in the population is positive or negative.
How can p-values from a two-sided test be converted for a one-sided test?
Divide the two-sided p-value by 2.
Why must one-sided tests be justified?
To avoid accusations of cherry-picking or cheating.
What is the line of best fit in scatterplots?
A line that represents the data’s pattern in a correlation.
How are correlations reported in scientific articles?
Often in scatterplots or correlation tables.
What does r(A, B) = r(B, A) mean?
Correlations are symmetrical; the order of variables does not change the result.
What does “Correlation ≠ Causation” mean?
Correlations show associations but not whether one variable causes the other.
What is an example of a spurious correlation?
Shoe size correlates with reading ability due to age, not a direct relationship.
What is the importance of time order in correlations?
It can sometimes exclude one variable as the causal factor (e.g., temperature rise before film crews arrive).
What are degrees of freedom (df)?
The number of observations that are free to vary when computing statistics.
How is df calculated in correlation tests?
df=n−2, where n
n is the sample size.
What is the difference between a two-sided and one-sided correlation test?
Two-sided tests for any difference, while one-sided tests for a specific direction.
What is the main use of scatterplots?
Visualizing the strength, direction, and shape of associations.
How are correlation coefficients visualized?
In scatterplots or tables with associated p-values.
What does p < 0.05 indicate in correlation tests?
The result is statistically significant.
How are correlation coefficients reported?
As r, with up to 3 decimal places, bounded between -1 and +1.
What is the role of p-values in reporting correlations?
They indicate whether the correlation is statistically significant.
How does practice improve correlation interpretation?
By identifying patterns in scatterplots and computing correlations.
