CHAPTER 4 Flashcards
when do two variable have an association?
- when there is a relationship between them
where should you plot the explanatory (independent) variable when drawing a scatter diagram?
- x axis
what is correlation?
- association between two variables that show increasing or decreasing trend
what is positive correlation?
- one variable increases when the other increases
what is negative correlation?
- when one variable decreases as the other increases
when is the correlation linear?
- when the points lie on or are near a straight line
what is a causal relationship?
- when a change in one variable directly causes a change in another variable, there is a causal relationship between them
- correlation does not necessarily imply a causal relationship
how to get a good fit when drawing a line of best fit:
- draw line through mean point
- sometimes written as (x, y)
both with lines on top
what is interpolation?
- using a line of best fit to estimate values within the data range
- usually reliable
what is extrapolation?
- using a line of best fit to estimate values outside the data range
- less reliable as they are further away from the range of data
equation of a line of best fit:
- y= ax + b
- a= gradient
- b= y intercept
what is the other name for the line of best fit?
- regression line
what does the gradient of a line of best fit represent?
- rate of increase of the response variable in relation to the explanatory variable
what does the y- intercept of a line of best fit represent?
- value of response variable when the explanatory variable is 0
what does spearman’s rank correlation coefficient measure?
- strength of correlation between two sets of data
what does it mean if r is close to 1, 0 or -1 (spearman’s or Pearson’s)?
- 1= strong positive correlation
- 0= no correlation
- -1= strong negative correlation
what does Pearsons product moment correlation coefficient measure?
- strength of linear correlation between two sets of data
when is Pearson’s product moment correlation coefficient suitable?
- when data shows linear correlation
when is spearman’s rank correlation coefficient suitable?
- when data shows non- linear correlation