PS125 Research & Statistical Methods Term 1 Part 1

Question 1

What is the Scientific Method?

Answer 1

Question
Hypothesis
Prediction
Experiment
Analysis
Interpretation

Question 2

Define population

Answer 2

Complete set of events of interest

Question 3

Define sample

Answer 3

Subset of a population

Question 4

Define parameter

Answer 4

Descriptive measure of a population

Question 5

Define random sample

Answer 5

Each number of population has equal chance of inclusion

Question 6

A good hypothesis should be

Answer 6

Be stated in declarative form
Posit a relationship between variables
Reflect a theory or a body of literature on which they are based
Be brief and to the point and
Be testable

Question 7

Features of Null Hypothesis

Answer 7

All things are equal or unrelated
No relationships between variables
No difference between groups
Often opposite of the research hypothesis

Question 8

True or false you NEVER accept the null hypothesis

Answer 8

True

Question 9

Define probability

Answer 9

The degree of confidence we have in stating that a particular outcome may not have occurred due to chance alone

Question 10

Give the formula for relative frequency

Answer 10

Frequency/N

Question 11

What is data?

Answer 11

Information in numeric form
Can represent almost anything
Meaningless without labels and context
Allows us to do mathematical and statistical analysis

Question 13

What features are part of a report?

Answer 13

Title
Abstract
Introduction
Method
Results - descriptive and inferential
Discussion
References

Question 14

What is the disadvantage of the Interquartile Range?

Answer 14

IQR discards lowest 25% and highest 25% of scores and then calculates the range

Question 15

Frequency distributions

Answer 15

Frequency distributions - Represent the number of occurrences of each value in each data set
E.g scores of 30 children on a reading test (N = 30)

Question 16

Histograms

Answer 16

Graph in which rectangles are used to represent frequencies of observations within specified intervals

Question 17

Define skewness

Answer 17

deviation from symmetry (same shape on both sides of centre)

Question 18

Rough draw a negatively, symmetric and positively skewed graph

Question 19

Unimodal vs bimodal

Answer 19

One mode vs two modes

Question 20

Define Kurtosis

Answer 20

Peakedness of distribution

Question 21

Rough draw a platykurtic graph vs a leptokurtic graph

Question 22

Standard deviation

Answer 22

How spread out data is around the mean
Average distance or deviation of each score from the mean
Standard deviation is the square root of each variance

Question 23

Define theory

Answer 23

Logically organised set of proposals that defines, explains and organises our knowledge about (many) significant aspects of behaviour

Question 24

Define hypothesis

Answer 24

A formally stated expectation about behaviour that defines the purpose and goals of a study

Question 25

Quasi-experiments

Answer 25

When we CAN’T manipulate our independent variable we have to use and existing or inherent characteristic
Can’t randomly assign people to conditions
Need to consider the wider context in interpreting data
May not be able to draw strong casual relationships

Question 26

Features of experiments

Answer 26

Can be tightly controlled
Reliable
Good internal vailidity
May lack ecological validity

Question 27

Features of Quasi-experiments

Answer 27

Hard to control
May have cohort effects
May lack internal validity
Good ecological validity

Question 28

Observational studies

Answer 28

Excellent ecological validity
Low demand effects
Poorly controlled
Ethically complex

Question 29

Correlational studies

Answer 29

NO IV
Looks for relationships, not differences

Question 30

When measuring a dependent variable you must be ______, _______ and _______

Answer 30

Valid, reliable and ethical

Question 31

Probability and Sampling Distributions

Answer 31

For continuous sampling distributions, probability is defined by area/regions under the curve
Total area under the curve adds up to 1
Area under the curve between any specified points represents the probability of obtaining a score with those points

Question 32

What is skew direction determined by?

Answer 32

Skew direction is determined by low frequency values and mean is ‘pulled’ in the direction of skew

Question 33

The Normal Distribution four defining characteristics

Answer 33

Symmetry
Unimodal
Limits of -∞and +∞
Continuous

Question 34

What two parameters does the normal distribution have?

Answer 34

The mean and standard deviation

Question 35

Central Limit Theory

Answer 35

CENTRAL LIMIT THEORY - if sample is large enough, sampling distribution of the mean will be normally distributed

Question 36

Type I error

Answer 36

Type I error is the error of rejecting the null hypothesis when its true (FALSE POSITIVE)
The Type I error rate lies in one tail
The two-tailed tests divide the type I error rate equally between two halfs

Question 37

Type II error

Answer 37

Type II error is the error of NOT rejecting the null hypothesis when its false (FALSE NEGATIVE)

Question 38

P-values and Type I error

Answer 38

P-values and Type I error are probabilities
Can interpret p-values in terms of Type I error for both one and two tailed tests
P-value can be considered the Type I error rate I would need to assume to reject the null hypothesis
For both one and two tailed tests if p-value is less than Type I error rate, reject the null hypothesis

Question 39

Features of different methods used

Answer 39

Observational studies
No manipulation done in the field
Surveys/questionnaires
No manipulation; just ask questions
Quasi-experiments
No direct manipulation uses pre-existing properties e.g men vs women
Experiments
An experiment involves the manipulation of one or more variables by an experimenter in order to determine the effect of this manipulation on another variable

Question 40

Within-participants study

Answer 40

Within-participants - ALL participants experience ALL conditions
Avoids contamination of individual differences
May need counterbalancing

Question 41

Between-participants study

Answer 41

Between-partcipants - Participants allocated to separate groups or conditions, groups are compared
Quasi-experiments
Avoids practice effects, allows naive participants
Requires more participants

Question 42

Define confound

Answer 42

An additional factor which could affect our results

Question 43

The special important case of normal distribution

Answer 43

Mean μ = 0 and Standard deviation σ = 1
Very unlikely that your null hypothesis is the standard normal distribution

Question 44

Z-score formula

Answer 44

sample mean - population mean/standard error of the mean

Question 45

Z-score features

Answer 45

Change the mean of a distribution to 0 and the standard deviation to 1
Equals the number of standard deviations above or below the mean

Question 46

T-test

Answer 46

Limited number of cases where population standard deviation is known
Problem: without standard deviation, the sampling distribution is defined (the two parameters are NEEDED to define a distribution)
Solution: estimate the population standard deviation from the sample deviation

Question 47

T-test formula

Answer 47

particular sample mean - population mean/standard error of sample means

Question 48

The Student’s T Distribution

Answer 48

The (Student’s) t distribution
Theoretical distribution
Symmetrical and bell-shaped
EXCEPT standard error not standard deviation
Standard deviation is estimated as we gather more samples we become more confident in our ability to estimate the standard deviation
T is not normally distributed, so we cannot use the SND tables

Question 49

The t-distribution parameter

Answer 49

t distribution has one parameter: degrees of freedom
Df number of individual scores that can vary without changing the sample mean
Df equals one less than the number of observations in the sample (N-1)
If df = ∞ then t distribution is identical to SND
Otherwise t distribution looks similar to SND but with slightly thicker tails

Question 50

If p-value < α value the null hypothesis is rejected

Question 51

Cohen’s d

Answer 51

d=0.2 represents a ‘small’ effect size
d=0.5 represents a ‘medium’ effect size
d=0.8 represents a ‘large’ effect size