What frequency distributions are most appropriate for nominal-level variables?
Raw and relative frequency
What frequency distributions are most appropriate for ordinal-level variables?
Raw, relative, and cumulative frequencies
What frequency distributions are most appropriate for interval/ratio-level variables?
Raw, relative, and cumulative frequencies along with class intervals
Is it appropriate to use a cumulative percent distribution with a survey on respondents preferred mode of transportation?
NO - because this is a nominal-level variable (can't be ranked)
It is appropriate to use a relative frequency distribution with data on the number of kids per household in a neighborhood?
YES - because this is an interval/ratio-level variable (can be ranked with clear distinction between intervals)
Is it appropriate to use a relative frequency distribution and a cumulative frequency distribution with data on how many kids of a student body make up each grade at a high school?
YES - because this is an ordinal-level variable (can be ranked)
What are the best or most appropriate ways to display nominal-level data?
Raw and/or relative frequency
Mode
Pie charts and bar charts
What are the best or most appropriate ways to display ordinal-level data?
Raw, relative, and cumulative frequencies can be used
Median and sometimes mode if you're interested in the most common answer
Bar charts and pie charts
What are the best or most appropriate ways to display interval/ratio-level data?
Raw, relative, and cumulative frequencies can be used along with class intervals
Mean and sometimes median and mode
Histograms
When should you use the mode of a data set?
Nominal-level variables
To quickly measure ordinal or interval/ratio-level variables
When you're interested in the most common answer
When should you use the median of a data set?
Ordinal-level variables
Interval/ratio data that's skewed
When you're interested in the 50th percentile
When should you use the mean of a data set?
Interval/ratio-level data that's relatively symmetrical
If you're doing even further statistical analysis
If a graph of income levels in America is 'positively skewed' what does that mean?
There are a couple of extremely large values 'pulling' the tail to the right
It means that the mean is greater than (>) the median
If a graph of education levels in America is 'negatively skewed' what does that mean?
There are a couple of extremely low values 'dragging' the tail down to the left
It means that the mean is less than (<) the median
When is the range of a distribution useful?
When you want to know the difference between the lowest and highest score
When is the IQR of a distribution useful?
When you want to find the middle 50% of a distribution
When you only care about the middle and 'average' values of a distribution since it ignores outliers either way
When is the variance of a distribution useful?
When you want to know how far each observation is from the mean of the distribution (how much each observation varies from the 'average')
When is the standard deviation of a distribution useful?
When you want to or need to return data to it's original form/metric