Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
49 Cards in this Set
- Front
- Back
3 Useful features of ration data
|
-shape of distribution
-central, average, or typical value (mean, median, mode) -variability |
|
Histograms (steps)
|
Display the shape of distributions
1. Divide data set into intervals, equal width 2. Count # of cases per bin 3. Bar height= # of cases within that bin |
|
Outliers
|
Unusual cases
|
|
Skewed to right/left/symmetric
|
fewer values on right vs. fewer values on left
|
|
Bimodal
|
2 peaks
|
|
Unimodal
|
1 peak
|
|
Mode
|
The most common value
- can also be used for ordinal or nominal data |
|
Mode strengths
|
East to understand & calculate
|
|
Mode weaknesses
|
-Not practical for a large # of categories
-Does not provide a lot of information |
|
Mean
|
The average, center of gravity as a distribution
|
|
Mean formula
|
M=Exi/N
m:mean E=sum of Xi N=total # of cases |
|
Strengths of mean
|
Simple to calculate & can tell you a lot about a distribution
|
|
Median
|
The value of the score that divides the date in half
|
|
Median strengths
|
Less sensitive to outliers than the mean
|
|
Range
|
Difference between minimum value and maximum value
|
|
Range strengths
|
Easy to calculate & understand
|
|
Range weaknesses
|
Only depends on large & small values
|
|
Interquartile Range
|
Difference between the upper & lower quartiles
|
|
Strengths of IQ range
|
Less influenced by largest & smallest values than the range
|
|
Weakness of IQ range
|
Hard to calculate & explain
|
|
Standard deviation (steps)
|
Average distance from the mean
1. calculate mean 2. Find deviation from mean for each value 3. Square these deviations 4. Sum squared deviations 5. Divide by n-1 6. Take square root of variance |
|
Variance
|
The standard deviation squared
|
|
Variance formula
|
S^2=E(Xi-M)^2/N-1
Xi: case value M: mean of cases E: sum of (Xi-M)2 N: total # of cases |
|
Standard deviation
|
s=Square root of s^2
(square root of the variance) |
|
Frequency curve
|
Taking a smooth curve along the top of a histogram to represent the data
|
|
Normal curve
|
A distribution whose frequency is bell shaped
- allows us to easily estimate the proportion of people within any given interval on the curve - we can also calculate what value corresponds to a given percentile and vice versa |
|
Standardized score (z-score) (formula)
|
# of standard deviations the observed value falls from the mean
Z score= (X-M)/S X: value M: mean S: standard deviation |
|
Positive standardized score
|
Means a score is above the mean
|
|
Negative standardized score
|
Means a score is below the mean
|
|
Standard normal curve
|
normal curve with a mean of 0 and SD of 1
|
|
Empirical rule
|
- 68% of values fall within 1 SD from the mean
- 95% of values fall within 2 SD from the mean - 99.7% of values fall within 3 SD from the mean |
|
To find an observed percentile:
|
1. Look up observed percentile & find corresponding standardized score
2. compute the observed value =mean- standardized score * standard deviation |
|
Reasons to use graphs
|
-Can convey a story relatively quickly
-They can also reveal things that would be difficult to see by looking at raw numbers -They can obscure meaning |
|
Common problems with graphs
|
-Labels
-Range of values displayed -Clutter |
|
Graphs used for categorical data
|
Pie chart
Bar graph Pictogram |
|
Graphs for ratio data
|
Histograms
Stem-and-leaf plots Line graphs Scatterplots |
|
Pie chart
|
Shows the percentage of cases in each category
-usually display 1 categorical variable |
|
Bar graphs
|
Also show the percentage of cases in each category, but can show more than 1 variable
-height of bar=percentage of cases in each category -sometimes designed to show percentages within a given category as well (all bars same height, split into 2 different categories) |
|
Pictogram
|
Like a bar chart, but uses a picture instead of a bar
-used in magazine articles, can be misleading |
|
Why can pictograms be misleading?
|
-should be avoided b/c they can be confusing
-height of bars represents value, but eye can be fooled into thinking area of figure represents the relevant value -can be hard to read |
|
Line graphs
|
Display lines indicating changes over time
|
|
Scatter plots
|
Display the relationship of 2 measurement variables
-show one "dot" for each pair of values for the 2 measurement variables |
|
Well designed pictures
|
-data should stand out clearly
-clear labeling: tile, source of data, axes or categories |
|
Deterministic relationships
|
1 variable perfectly predicts another
|
|
Statistical (probabilistic) relationships
|
There is a relationship but it is not precise
|
|
Correlation
|
positive, negative, no relationship
-express relationships with a single # |
|
Correlation coefficient (Pearson Product-Moment correlation)
|
Measures strength of relationship between 2 measurement variables
|
|
Pearson product-moment correlation
|
Range between -1 and 1
-correlation of 1= perfect linear relationships -0= no linear relationship |
|
Correlation coefficient
|
Calculated based on the formula for z-scores
r=EZxZy/n |