HL Psychology Internal Assessment

1
HL Psychology Internal Assessment

• Inferential Statistics

2
What you should know after this PowerPoint

• A concise review of descriptive statistics
• Differences between descriptive and inferential
  statistics.
• Why we use inferential statistics in psychology
• How to properly choose an inferential statistics
  test.
• How to distinguish between various types of data.
• How to test for statistical significance.

3
Descriptive statistics provide for..

• Measure of central tendency
• Gives a typical value for the data set
• Tells you where the middle of the data set is
• Measure of dispersion
• Indicates how the data are spread out
• Tells you what the rest of the data are

4
Descriptive Statistics

• The aim of descriptive statistics is to give an
  accurate summary of the data
• The wrong choice of statistic gives a distorted
  picture of the data
• This can lead to the wrong conclusions being
  drawn from the data
• Each measure of CT and D has its advantages and
  disadvantages

www.psychlotron.org.uk
5
Measures of Central Tendency

• The mean total scores divided by the number of
  scores
• Adv it uses all the values in the set, so is
  most sensitive to variations in the data
• Dis it can be artificially raised or lowered by
  an extreme value, or by skewed data
• Use it when the data are normally distributed,
  unskewed and there are no outliers

www.psychlotron.org.uk
6
Measures of Central Tendency

• The median the middle score in a range
• What is the median 2,3,3,4,4,4,4,5,5,6,42?
• Adv it is based on the order of the data, not
  their actual values, so not distorted by extreme
  values
• Dis however, this makes it less sensitive to
  variations in the data
• Use it when you cant use the mean because of
  skew, outliers etc.

www.psychlotron.org.uk
7
Measures of Central Tendency

• The mode -most frequently occurring value
• Adv its the only measure suitable for
  summarising category/frequency data
• Dis for many data sets there is no modal value,
  or their may be several
• Use when dealing with frequency data, and/or
  where there is a clear modal value in the set

www.psychlotron.org.uk
8
Calculate.

• A psychologist has obtained the following scores.
  Answer the questions below.
• 8 1 5 5 2 7 1 1 1 4 6 8
  9 9
• The range of these scores is _____________________
  _____
• The mean of these scores is ______________________
  ____
• The mode of these scores is ______________________
  ____
• The median is ____________________________________
  __

9
Measures of dispersion

• Range-difference between the smallest and largest
  value Ex 3,4,7,7,8,9,12,4,17,17,18 18-3 16
• Although quick and easy to calculate it is
  distorted by extreme values

10
Standard Deviation

• Standard deviation a measure of the spread of
  scores around the mean
• It is the most sensitive measure of dispersion
  using all available data. It can be used to
  relate the sample data to the populations
  parameters.

11
SD formula

• Sum of all participant scores divided by the no
  of participants mean
• Subtract the mean from each score
• Square each of these scores
• Total the squared scores
• Divide by one less than the total participants.
  This is the variance
• Take the square root of the variance.

12
Work out the SD.

• Scores 13,6,10,15,10,15,5,9,10,13,6,11,7

13
Graphs

• Bar chart Shows data for categories that the
  researcher is interested in comparing

14
Histogram

• Shows data for all categories even those with
  zero value

15
Frequency polygon/line graph

• Shows two sets of data on one graph

16
Pie charts

• Show the proportion of all scores gained by
  various categories

17
Inferential Statistics

• HL IA ONLY

18
Inferential Statistics

• With inferential statistics, you are trying to
  reach conclusions that extend beyond the
  immediate data alone. For instance, we use
  inferential statistics to try to infer from the
  sample data what the population might think.
• Or, we use inferential statistics to make
  judgments of the probability that an observed
  difference between groups is a significant one or
  one that might have happened by chance in this
  study.

19
Inferential Statistics

• Thus, we use inferential statistics to make
  inferences from our data to more general
  conditions we use descriptive statistics simply
  to describe what's going on in our data.

20
What you are bring asked to do (HL IA).

• An appropriate inferential statistical test has
  been chosen and explicitly justified. Results of
  the inferential test is accurately stated.
• The null hypothesis has been accepted or rejected
  according to the results of the statistical test.
  A statement of statistical significance is
  appropriate and clear.

21
What you are bring asked to do (HL IA).

• The information you have obtained from
  participants takes the form of raw data. This
  should go into the appendices, and you should use
  your results to calculate descriptive statistics
  appropriate to your to data.
• The test you choose is dependent on the level of
  measurement of your data and whether you used
  independent samples or repeated measures.

22
Levels of Measurement

• Nominal-frequency headcount things can only
  belong to one category ex the no of students
  wearing yellow shirts.
• Ordinal data which is ranked or put in order. It
  is not known what the interval between each rank
  is ex 1st,2nd,3rd time in a swimming trial
• Interval/ratio- measurement on a scale where the
  intervals are known and equal (ratio has a true
  zero point interval can move into negs. Ex of
  ratio is time in secs.

23
Levels of data nominal

• Which newspaper paper do you read regularly?
• We can put these into categories.

24
Levels of Data ordinal

• What grade did you get for each of your
  portfolio?
• These can be put in order highest to lowest

25
Levels of data interval

• How quick is your reaction time?
• We can measure and compare the exact time because
  the intervals on the ruler are equal.

26
(No Transcript)
27
Inferential tests

• Provide a calculated value based on the results
  of the investigation
• This value is then compared to a critical value
  (statistical tables) to determine if the results
  are significant
• In chi square, sign test, spearmans rho the
  calculated value must exceed the critical value.

28
(No Transcript)
29
Choosing an inferential test

• Nominal data and independent measures design
  Chi square test
• Ordinal data and independent measures design
  Mann Whitney U
• Interval and ratio data and independent measures
  design Unrelated T-test
• Nominal data and repeated measures design Sign
  test
• Ordinal data and repeated measures design
  Wilcoxon test
• Interval or ratio data and repeated measures
  design related T-test
• More info http//hs-psychology-ibhl.ism-online.o
  rg/files/2011/09/Choosing-an-inferential.pdf

30
A directional hypothesis

• Very often, we state before we test the
  hypothesis in which direction of the results will
  fall. Our hypothesis is usually directional
  (meaning we are predicting an increase or
  decrease in a time or score)and the appropriate
  statistical test of the hypothesis is called
  one-tailed.
• Once you have collected the data. Decide which
  test you need to administer. Only one person in
  your group needs to work out the mathematics.

31
Using Tests of Significance The General
procedure

• Choose appropriate statistical test
• Calculate statistical test
• Compare the test with the critical values. These
  can be found in the back of the Research methods
  text book, or mathematics statistic books, or
  online.
• Decide which side of the critical value your
  result is on.
• Report the decision.

32
Inferential statistics- indicating how
significant results are.

• A significant result is one where there is a low
  probability that chance factors were responsible
  for observed difference
• 5 level of significance, in psychology, is
  acceptable (P is less than 0.05)
• There is less than a 5 likelihood that the
  difference was due to chance.

33
Key Terms you will need to look up and define.

• Critical value
• Degrees of freedom
• P value/level
• Significance
• One-Tailed Test
• Two-Tailed Test
• Type 1 error
• Type 2 error
• Interval
• Ordinal
• Nominal