Research Methods
- Created by: Beverley
- Created on: 06-01-16 17:52
What is quantitative research?
Deals with numerical values
1 of 82
What does good quantitative research look like?
Generalisability, Validity, Reliability and Replicability
2 of 82
What is qualitative research?
Descriptions and data that can be observed but not measured
3 of 82
What is inductive research?
bottom up approach (begins with specific observations to broader generalisations and theories)
4 of 82
What is deductive research?
top up approach (begins with broad spectrum of information and work their way down to a specific conclusion)
5 of 82
What is pure research?
Exploratory research with no practical use in mind
6 of 82
What is applied research?
Solve practical questions and problems
7 of 82
What is primary research?
Research you have done yourself
8 of 82
What is secondary research?
Existing research
9 of 82
What is desk-based research?
Collecting data from existing resources
10 of 82
What is Laboratory research?
Research conducted in a room or building that is equipped for scientific experimentation or research
11 of 82
What is field research?
Research conducted outside of a lab experiment
12 of 82
Define a hypothesis
It is a claim / prediction / informed question based on theory of prior research
13 of 82
Different types of hypothesis
One tailed and Two tailed hypothesis
14 of 82
Define a one-tailed hypothesis
A hypothesis that predicts the particular direction of an effect (e.g. greater)
15 of 82
Define a two-tailed hypothesis
A hypothesis that predicts the particular effect but not the direction (e.g. between)
16 of 82
What 4 types of sampling methods are suitable for quantitative research?
Random, Stratified, Snowball and Convenience/Opportunity
17 of 82
Define random sampling
everyone in the population has an equal likelihood of selection (e.g. through a number generator)
18 of 82
Define stratified sampling
The population is sub-divided by known strata and participants are sampled randomly from within each stratum
19 of 82
Define snowball sampling
Population members act as an agent of the researcher by sampling from colleagues, friends or associates (useful when the population required is inaccessible to the researcher)
20 of 82
Define convenience/opportunity sampling
recruitment of participants who happen to be available
21 of 82
What is a variable?
They are anything that is subject to variation
22 of 82
What are the different types of variables?
Independent, Dependent and Extraneous Variable
23 of 82
What is an independent variable?
The input that researchers manipulate which affects the dependent variable (DV)
24 of 82
What is a dependent variable?
The output that is affected by the independent variable (IV)
25 of 82
What is an extraneous variable?
Other variables that might affect the dependent variable (DV) - e.g. different materials for each condition
26 of 82
Define Random Error
Variables which might affect a participants in the same way across both conditions of the experiment (e.g. how your feeling, stuffy room and noises)
27 of 82
Define Constant Error
(Systematic) affect one condition more than another (e.g. materials different between groups)
28 of 82
What is internal validity?
The extent to which we can be sure that changes in the DV are due to changes in the IV
29 of 82
What is external validity?
The extent to which we can generalise findings from the situation the researcher conducted to real-life situations (ecological validity) AND the extent to which we can generalise the findings from participants to the wider population
30 of 82
What does is mean if we have high ecological validity?
It means we can generalise the findings to a real-life situation
31 of 82
Name the types of experimental design
Laboratory, Quasi, Field and Natural experiments
32 of 82
How do true and quasi experiments differ?
Quasi experiments have no control over participants (allocated subjects such as males and females)
33 of 82
What are the advantages of a field/natural experiment?
No demand characteristics and high ecological validity
34 of 82
What are the disadvantages of a field/natural experiment?
It lacks ethics and consent, hard to replicate and generalise, variables are hard to measure and extraneous variables cannot be controlled
35 of 82
What are the advantages of a laboratory experiment?
Ability to control the IV, easily replicable, reduces the chance of extraneous variables and able to control the conditions
36 of 82
What are the disadvantages of a laboratory experiment?
The environment is too heavily controlled, researcher bias, low ecologial validity and participants may try to figure out what researchers are testing (Horthorne effect)
37 of 82
What is the independent / unrelated / between subjects measure?
Testing separate groups of people, each group is tested in a different condition
38 of 82
What is the repeated / related / within subjects measure?
Testing the same group of people in different conditions, the same people are used repeatedly
39 of 82
What does matched pairs mean?
Testing separate groups of people - each member of one group is the same age, sex or social background as a member of the other group
40 of 82
What are the levels of measurement?
Nominal, Ordinal, Interval and Ratio
41 of 82
What is nominal level data?
Participants are placed in categories (e.g. eye colour)
42 of 82
What is ordinal level data?
The order of a group of numbers but there's no accurate scale (e.g. position in a race)
43 of 82
What is interval level data?
Continuous scale whereby the distance between points is known (e.g. temperature, bank balance) - there is no fixed zero
44 of 82
What is ratio level data?
Continuous scale with equal intervals and a true zero (e.g. height, weight, distance, blood pressure)
45 of 82
Which levels of measurement provide the most useful quantitative data?
Interval and Ratio
46 of 82
What are measures of central tendency?
A numerical value that represents the most typical number in a group; it is somewhere in the middle. There are 3 different measures: Mean, Median and Mode
47 of 82
What is subjective probability?
When we talk about the probability of something happening (e.g. how likely England will win the world cup)
48 of 82
What is logical probability?
It is a statement of a ratio of favourable cases to total number of equally likely cases (e.g. likelihood of picking an ace from a pack of cards)
49 of 82
What is empirical probability?
The probability of an event occurring based on historical events
50 of 82
What is the null hypothesis?
The statement of no (null) difference between the scores in the different conditions of variable tested (If we accept the results occurred by chance, we accept the null hypothesis)
51 of 82
What is the alternative/experimental/working hypothesis?
It predicts there will be a difference between the scores of the two groups (If the experimental hypothesis is upheld, then we can reject the null hypothesis)
52 of 82
What value do we use when to express statistical significance?
p < .05
53 of 82
What does a type I error mean?
Where we conclude that the IV was responsible for an affect on the DV when it was not something else caused it - False Positive
54 of 82
What does a type II error mean?
If we fail to achieve a p < .05 we will conclude that the IV had no major effect on the DV - False Negative
55 of 82
What is normal distribution?
Particular way in which individual scores/values are dispersed or spread
56 of 82
What are the pre-requisites for normal distribution?
(1) Continuous data (2) Fine levels of measurement - no limit to how finely we can measure things (3) Large sample size
57 of 82
Define skewness
A lack of symmetry around the central vertical line (mean, median, mode)
58 of 82
Normal distribution has a skew of 0. What is considered acceptable skew?
-1 to +1
59 of 82
Define Kurtosis
Relates to how 'pointy' or 'flat' the distribution of scores is
60 of 82
Normal distribution has a kurtosis of 0. What is considered acceptable kurtosis?
-2 to +2
61 of 82
What % of data is 1 standard deviation (1 SD)
68% of data
62 of 82
What % of data is 2 standard deviation (2 SD)
95% of data
63 of 82
What % of data is 3 standard deviation (3 SD)
99.7% of data
64 of 82
What is a z-score?
Is a number of standard deviations a particular raw score is away from the mean
65 of 82
What is the function of a z-score?
To compare different people on a test, compare the same person on different tests and if you know the z-scores you can calculate the % of cases above and below the score
66 of 82
What is a t-test?
Looking at differences between groups of people or data, by manipulating what happens we can make casual inferences
67 of 82
What is the purpose of a t-test?
Establish whether the means of two samples are statistically different from each other and to compare with r for our correlations
68 of 82
What is the observed value of a t-test?
t
69 of 82
What are the 3 parametric assumptions?
(1) The DV is continuous - interval/ratio level data (2) There is a normal distribution - no skew or kurtosis (3) There is homogeneity of variance where the variance between groups is approx the same
70 of 82
What test is used to test homogeneity of variance?
The Levene's test
71 of 82
Do you want the Levene's test to be significant?
No
72 of 82
What are degrees of freedom?
Tells us how many bits of data are free to vary in our data set and how much data was used to calculate a particular statistic
73 of 82
How to calculate degrees of freedom?
df = number in a group - 1
74 of 82
What does the effect size tell us?
It tells us how big the effect we've found may be
75 of 82
Effect size is measured as a value between...
0 and 1
76 of 82
When are non-parametric tests used?
When they don't meet the 3 parametric assumptions
77 of 82
What is the non-parametric test equivalent of a between t-test?
Mann-Whitney U Test (U)
78 of 82
What is the non-parametric test equivalent of a within t-test?
Wilcoxon Signed Ranked Test (T or Z)
79 of 82
What is a Chi-square test?
A test of association between data for 2 categorical variables (observed vs. expected)
80 of 82
What are the 3 assumptions of a chi-square test?
(1) Nominal level data (2) All the data are in a form of frequencies (3) All data is independent - can only be put into one category
81 of 82
How to calculate the degrees of freedom for a chi-square test?
df = (number of rows - 1) * (number of columns - 1)
82 of 82
Other cards in this set
Card 2
Front
What does good quantitative research look like?
Back
Generalisability, Validity, Reliability and Replicability
Card 3
Front
What is qualitative research?
Back
Card 4
Front
What is inductive research?
Back
Card 5
Front
What is deductive research?
Back
Related discussions on The Student Room
- 5 note-taking methods all students should know »
- How to start revising »
- How do you guys remember all the case studies + details for A level »
- MA Big Data at KCL vs MSc Social Research Methods LSE »
- Math question »
- access to HE Introduction to Psychology unit.. HELP NEEDED »
- Aerosol_man's Laptop recommendation thread »
- A-level geography NEA »
- UAL didn’t accept offer yet »
- angbrads's Laptop recommendation thread »
Similar Psychology resources:
0.0 / 5
0.0 / 5
0.0 / 5
5.0 / 5 based on 1 rating
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
0.0 / 5
Comments
Report
Report
Report