key things to remember for exam tomorrow Flashcards
Variance
it measures how much each data point differs from the mean
gives info about the variability within the sample population.
inferential statistics
infer info and draw conclusions
-compare different conditions in an experiment
-draw conclusions from the experiment to the total population
descriptive statistics (mean, range, mode, median)
-describe the overall sample population but on their own they can be limited
-summary of data
standard deviation
tells us whether there is a lot of difference between the different data points/participants
low SD- Ps are all similar
high SD- some Ps are very different (anomalous)
Normal distribution fun facts
Curve never touches the x-axis and extends to the left and right forever!
always will be one person who is extreme
COMPLETELY SYMMETRICAL: In the normal curve, the mean, median, and mode are all the same (50% data is below and above the mean)
-fixed percentages of scores fall between points given by the standard deviation
Measures of Central tendency
mean
median
mode
measures of variability
variance
range
standard dev
always report central tendency and measures of variability at the same time
positive scew
negative scew
left- mean is lower
right- mean is higher
z score
a numerical measurement that describes a values relationship to the mean of a groups values and is measured in terms of standard deviations from the mean
z score summary
➢A z-score indicates how many SDs you are
away from the mean.
➢If a z-score is equal to 0, it is actually on
the mean.
➢Positive z-scores: the raw score is higher
than the mean average.
➢Negative z-scores: the raw score is below
the mean average
Rescaling
Can compare,
combine or average
the scores because
they come from a
distribution with the
same mean and SD
Probabilities vs. Percentages
mean the same thing
Positive z score- As the score moves further from the mean (i.e. the z-score increases), the larger portion gets bigger, and the smaller portion decreases in size
Z-scores also allow us to
calculate percentiles
levels of measurement
nominal
ordinal
interval
ratio
ratio level of data
The highest level of measurement
Has magnitude, equal intervals and
absolute zero
e.g., the Kelvin scale, weight, length,
time?
Histograms: visualise quantitative data
Bar charts: categorical variables
One sample t-test
Tests whether a population mean is
significantly different from your
hypothesized value.
Note – this t-test is NOT used very often
Indep t test (unpaired)
Compares two samples to each other.
INDEPENDENT (two different
samples)
Example:
Intervention study (two groups take part
in 2 different interventions)
Repeated measures t-test (or paired)
Compares two samples from data that
is related
REPEATED MEASURES (same
sample)
Example:
A pretest score and post test score (from
same participant)
Main diffs between t tests
Assumptions
Data structured in SPSS differently
(depending on design)
Remember: 1 participant per row
How to perform analysis in SPSS
Assumptions- one sample test
For a valid test, we need data values that are:
Independent (values are not related to one another).
Continuous (i.e., interval or ratio level)
Obtained via a simple random sample from the
population.
The population is assumed to be normally
distributed.
Homogeneity of variances (i.e., variances
approximately equal in both the sample and
population)
No outliers
H1 = two-tailed
