Research Methods AS L13 - 15 (graphs, significance levels, sign test) Flashcards
How can the results that appear in tables be described as?
- Descriptive statistics (Measures of central tendency/dispersion)
- Not raw scores
Measures of central tendency:
- Mean
- Mode
- Median
Measures of Dispersion:
- Range
- Standard deviation
Types of graphs:
- Bar chart
- Histogram
- Line graph
- Scattergraph
- Distribution Curve
Bar chart:
- Represents frequency of each item
- Space between bars
- Nominal data
Histogram:
- Area under bars must equal to frequency represented
- No gaps between bars
- Ordinal/interval data
Line graph:
- Connected by lines to show change over time
- IV on x axis, DV on y axis
- Ordinal/interval data
Scattergraph:
- Relationship between co-variables
- Closer the points, stronger the correlation
What shape should the frequency of the measurements reflect in most data sets and what is this known as?
- Bell-shaped curve
- Normal distribution curve
Do the ends of a normal distribution curve touch x axis and why is this the case?
- No
- More extreme scores are always theoretically possible
What are the names of asymmetrical normal distribution curves?
Skewed distributions
Descriptive stats:
Measures of central tendency/dispersion
Inferential stats:
Generalising one result to the population (inference)
Level of Statistical Significance:
- The level at which the decision is made to reject the null hypothesis in favour of the experimental hypothesis
- States how sure we can be that IV will have an effect on DV and it is not due to chance
Chance:
Something that has no real cause and just occurs
In which situation would the experimental hypothesis be accepted and the null hypothesis rejected and what is this called?
- If there is a real difference between both sets of data
- Results are significant
Probability:
Numerical measure determining whether our results are due to chance or whether there is a real difference between the sets of data
What is the standard level of significance used in psych and why is this number used?
- p < 0.05 (5% results are due to chance , 95% certainty there is real difference)
Reasons for use: - Not too strict or too lenient, but is a middle and fair value of significance
- Minimises chances of making Type 1 or Type 2 error
In psychology, what is the highest and lowest level of significance possible and what numbers are they?
- Highest, most lenient: p < 0.10 (10% results are due to chance , 90% certainty there is real difference)
- Lowest, most strict: p < 0.01 (1% results are due to chance, 99% certainty there is real difference)
Often when is p < 0.01 used?
When research findings are critical eg. testing effect of drug
Why is a sign test used?
- Looking for difference between data
- Looking at paired/related data –> could come from repeated measures/matched pairs design
- Nominal data
Steps of sign test:
1) State hypothesis (either one or two-tailed)
2) Subtract one column from another and work out the difference
3) Note down the sign of difference (positive/negative)
4) Count the number of positive and negative signs
5) Choose the lower value to be S value
6) Find out how many pps there are (ignore any blanks/differences of zero)
7) Find critical value in table using N, probability and whether it is one/two tailed
8) If s ≤ critical value, results are significant
How would the results of a sign test be written out?
The results of the experiment are significant. The calculated value for S is lower than/equal to the critical value of …. when p < 0.05 (usually unless question says otherwise) under a two-tailed hypothesis.
