Descriptive Statistics - 1 Flashcards
Frequency
Number of times one value occurs.
Frequency Distribution
Write down all the values in order. Count the frequency with which each value occurs.
Bar-charts and Histograms
Bar-charts (with gaps) are similar to histograms but the value labels become individual categories as it is not a continuous variable.
What are the measures of central tendency?
1) Mean
2) Median
3) Mode
Mean
Total scores and divide by number of scores.
Median
Put scores in order; pick the one in the middle. More than one median; add together and divide by two.
Mode
The score that happens most frequently. ‘Multimodal’ distribution; more than one mode.
What are the measures of dispersion?
1) Range
2) Standard deviation
Range
Take away the lowest score from the highest score.
Standard deviation
It measures how widely spread the values are around the mean. Unlike the range, it take all the values into account.
- If the data values are close to the mean, the SD is small.
- If the data values are far from the mean, the SD is large.
- If all the values are the same, the SD is zero.
You will not need to work out SD in the exam, only demonstrate what it tells us about the data, eg, more individual differences in condition with higher variance (larger SD).
Correlational Analysis
In a correlational study, the variables have to be identified and given a numerical value. The correlational relationship is identified and demonstrated using statistical techniques. The direction and strength of the correlation is measured.
Direction
- Positive correlation; as one variable increases, the other one also increases (eg. height and weight).
- Negative correlation; as one variable increases, the other variable decreases (eg. age and memory).
Graphs features
When drawing graphs, make sure:
- Detailed title
- Axes clearly labelled
- Appropriate choice of scale
- Correct plotting of points or bars
When should mean be used?
Assumes at least interval level of data (parametric).
Advantages of mean
Most sensitive measure of central tendency, taking all scores into account.
Disadvantages of mean
Can be distorted by extreme scores.
When should median be used?
Generally used with ordinal level data.
Advantages of median
Unaffected by extreme scores. Better than the mean if extreme scores exist.
Disadvantages of median
It only takes into account 1 or 2 scores.
When should mode be used?
Used with nominal data.
Advantages of mode
Similar to the median, unaffected by extreme scores.
Disadvantages of mode
Can be dramatically affected by change in 1 score.
When should range be used?
Basic measure of variation when the data is consistent.
Advantages of range
Easy to calculate.
Disadvantages of range
Easily distorted by extreme scores.
When should standard deviation be used?
More sensitive measure of dispersion when data is inconsistent.
Advantages of standard deviation
Takes into account all scores.
Disadvantages of standard deviation
More difficult to calculate than the range.
When should correlation be used?
When variables cannot be manipulated.
Advantages of correlation
Can statistically analyse situations that could not be manipulated by the researcher for practical/ethical reasons.
Disadvantages of correlation
Does not establish cause and effect.