PSYC1001 Lecture Notes - Lecture 18: Null Hypothesis, Statistical Inference, Sampling Distribution
2 Jul 2018
School
Department
Course
Professor
19/04/2018 Variability and Inferential Statistics
Frequency distribution graphs:
Each score is on the axis
How frequent a score is depending on the height of the line above the score
Most have a big bump in the middle, that’s because the most common scores are in the
middle (the top of the big bump is called the mode)
Inferential statistics – we take a sample from a population and we run our study on that
sample – when we obtain the findings – we must ask: Can we infer from the findings in our
sample, that there are differences in the population? – we ask because sampling variability
alone can cause effects in a sample
Sample = a selection from the entire collection you are interested in; properties of these
scores are called statistics and use Latin (normal) letters e.g. Mean = M
Population = the entire collection in which you are interested in; properties of these scores
are called parameters and use Greek letters e.g. Mean = µ
Sampling distribution – obtaining a first mean from the initial sample and then a second
mean which differs from the first because it comes from a different sample from the
population – continually repeating
Sampling distribution is a hypothetical distribution based on a hypothetical set of sample
means; the standard deviation of these means is called standard error; the shape of a
distribution of a million-hypothetical sample means tends to be normal regardless of the
distribution of the raw scores
Normal distribution shows estimation of the probability of obtaining a sample mean –
normality
How likely is it that we would obtain a sample mean of this value if our null is true?
In psych, usually decide something is interesting is happening when there is less than 5%
chance that we obtained the result we did by chance – work this out by calculating a
distance from the hypothesised null mean in standard error units
The bigger these numbers the less likely it is that the sample mean was sampled from a
population where nothing was happening
The actual probability of obtaining a sample value of a magnitude is expressed as a p-value
P values vary between 0 and 1 and express a probability – it is basically the probability that
we obtained the sample mean we did when the null is true by chance alone (not exact
definition but will do)
Researchers want to obtain low p-values – suggests their null hypothesis is false and they
have actually found something
In psych, p-values less than 0.05 we would normally reject the null hypothesis, if its bigger
we would normally retain the null hypothesis
A small p-value (0.002, 0.01) means: it is highly improbable that we obtained our result by
chance alone
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
How frequent a score is depending on the height of the line above the score. Most have a big bump in the middle, that"s because the most common scores are in the middle (the top of the big bump is called the mode) We ask because sampling variability alone can cause effects in a sample. Sample = a selection from the entire collection you are interested in; properties of these scores are called statistics and use latin (normal) letters e. g. mean = m. Population = the entire collection in which you are interested in; properties of these scores are called parameters and use greek letters e. g. mean = . Sampling distribution obtaining a first mean from the initial sample and then a second mean which differs from the first because it comes from a different sample from the population continually repeating. Normal distribution shows estimation of the probability of obtaining a sample mean normality.
