Review of Basic Tests of Difference and Relationship Flashcards
Correlation also known as?
Bivariate and relationships tests.
What is a correlation used for?
To examine the degree of association or relationship between variables.
Correlation definition?
A numerical coefficient that indicates the extent to which 2 variables are related.
The coefficient/number that represents the correlation is always between…?
-1 and +1.
What do correlation coefficients provide?
Information about the strength of a relationship.
What does a correlation coefficient of 0 indicate?
That the variables are uncorrelated (no relationship).
On a graph, which variable is better placed on the Y axis?
The dependent variable.
Ellipse?
A plane curve surrounding two focal points.
The less the variables are scattered on a scatterplot…
…the greater the relationship.
Key aspects to look at of an ellipse?
Incline and width.
What % confidence do you need to have in order to determine that there’s a relationship between different variables?
95%.
R value?
Correlation coefficient.
P value?
Tells you about ‘confidence’.
When should you produce a scatterplot of a relationship?
When can you include a trend line?
If it’s an important finding.
If the relationship is significant.
The parametric statistical procedure to test for a relationship is what?
Pearson’s product moment correlation.
Two main variables from Pearson’s product moment correlation test?
R value (correlation coefficient) P value (probability of error associated with accepting the alternative hypothesis/HA)
Which numbers is the P-value between?
0 and 1.
In order to accept the alternative hypothesis, what does the P value need to equate to? (what is the critical P value)
P = 0.05 or less (Critical P value = 0.05)
P = 0.59
What does this mean?
59% chance of error if you accept the alternative hypothesis.
What do correlation coefficients provide an indication of?
An indication of the linear relationship between variables.
When will the correlation coefficient underestimate the strength of a relationship?
When variables are related in a curvilinear way.
What is a curvilinear relationship?
Due to this existing, what is a good idea to draw?
A type of relationship between two variables where as one variable increases, so does the other variable, but only up to a certain point, after which, as one variable continues to increase, the other decreases.
Therefore it is a good idea to draw a scatter plot when conducting a correlation.
Casual relationship?
What is it also referred to as?
When one variable has a direct influence on another.
Referred to as cause and effect.
A significant relationship between two variables does not mean…
…that there is a casual relationship.
What is ‘the third variable problem’?
It could be true that A causes B, B causes A. But another variable C could also influence/’cause’ A and B.
T-test?
Tells you how significant the differences between groups are. It lets you know if those differences could have happened by chance.
Ratio scale?
- Has an absolute zero.
- Has orders and equally spaced units.
- No negative value.
- Values can be added, subtracted, multiplied and divided
- It has ratio scale units which allow unit conversion.
Unit conversion?
E.g. 12inches = 1foot.
Interval scale? Example?
Same characteristics as ratio scale but no true zero is used e.g. celsius can go below 0 and is therefore an example of interval scale however KG, for example, cannot.
Examples of non-parametric tests?
- Mann-Whitney U test
- Wilcoxon signed rank test
- Kruskal-Wallis test
- Chi-squared test
Kurtosis?
The sharpness of the peak of a frequency-distribution curve.
Skewness?
Skewness refers to a distortion or asymmetry that deviates from the symmetrical bell curve, or normal distribution, in a set of data.
Examples of parametric tests?
- Independent samples T-test
- Paired samples T-test
- One way ANOVA
Two tailed hypothesis?
A two-tailed hypothesis test is designed to show whether the sample mean is significantly greater than and significantly less than the mean of a population.
Type 1 error?
Type 2 error?
- The mistaken rejection of a null hypothesis as the result of a test procedure.
- A type II error is also known as a false negative and occurs when a researcher fails to reject a null hypothesis which is really false.
Statistical inference?
Process of drawing conclusions about a population based upon the sample data.
4 key features of a hypothesis?
- ) Difference or relationship?
- ) Dependent variable
- ) Independent variable and the levels of the IV
- ) Null and alternate versions
When to use an independent samples t-test?
If you are interested in identifying differences between groups of individuals.
On SPSS, sig. value = ?
P value.
What test assumes homogenity of variance?
How do we know this assumption is met?
- Levene’s test.
- No peculiar F or P values.
