year 1 recap

Independent variable

A variable which is manipulated by the researcher to test the effect it has

Dependent variable

measures the effect that manipulating another variable has

Variance

An estimate of average variability (spread) of a set of data.

Confounding variable

A variable that effects the outcome being measured as well as the independent variable

Degrees of freedom

The sample size- 1

The mean

The sum of all scores divided by the number of scores

the value from which scores deviate the least

a hypothetical value that doesn’t have to be a value in the dataset

Quasi experimental deign

An experimental design which lacks random assignment because groups are pre-existing or assigned based on criteria.

It’s purpose is to test potential causal relationships

deviation

The difference between the mean and the actual data point.

How to calculate deviation?

Taking each score and subtracting the mean from it

Sum of squared errors

Where you add the squared deviations together to find out the total squared error

Standard deviation

The dispersion of the dataset in relation to its mean

T-test

Determines whether there’s a statistically significant difference between the mean scores of two sets of values

When normality is assumed…

Use a:

Within samples t-test - 1group

Between samples t-test - 2 groups

When normality cannot be assumed use a…

Non- parametric equivalent such as mann-u-Whitney t test or wilcoxon w t-test

What type of data is used for parametric t-test?

Interval or ratio- data is continuous

What is the homogeneity of variance?

Data varies around the respective mean

What type of data is used when normality cannot be assumed?

Data is ranked

A normal distribution suggests that….

The mean, median and mode all fall around the same datapoint, being symmetrical

955 of data falls within the…

1.96 threshold

Confidence intervals

The boundaries in which we think the true mean of the population lies.

What does a small confidence interval suggest?

The sample mean Is very close to the true mean of the population

The result of the t-test depends on 2 factors-

The size of the difference between the means, the variability in the data

Type 1 error

Type 2 error

When we believe there is no effect in the population when in reality here is

Why p values are not enough?

P values are tied to sample sizes

They are not standardised

Different sample sizes will yield different p values for the same effect magnitude

Effect sizes

A standardise measure of the size of an effect, being comparable across studies

Different ways to measure effect size

Cohen’s d

EA squared n2

Partial ETA squared n2p

Z score

Indicates the number of standard deviations a score is from the mean

Partial eta squared is a measure of..

Skegness

4 aspects of normally distributed data…

• the distribution is symmetrical around the mean

• The values of mean, mode, median is the same.

• There are scores both above and below 2 SD.

• 68% of scores fall within 1 stSD

A frequency distribution in which low scores are most frequent (i.e. bars on the graph are highest on the left hand side) is said to be

Positively skewed

frequency distribution in which there are too many scores at the extremes of the distribution said to be

Platykurtic

Standard error is a measure of…

The variability of sample estimates of a parameter

What is the relationship between sample size and standard error of the mean?

Standard error decreases as sample size increases