Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

PSYCHOLOGY AS > Research Methods: Part 2 RM In Action > Flashcards

Research Methods: Part 2 RM In Action Flashcards

You may prefer our related Brainscape-certified flashcards:
Psychology 101 flashcards GRE® Psychology flashcards AP® Psychology flashcards
1
Q

Describe a research Aim.

A

Research aims state why experimenters are carrying out the project.
May begin with “to find out whether… Or to investigate the relationship between”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Describe a hypothesis.

A

A clear testable statement of the researchers prediction of what will happen.
Called the experimental hypothesis when the method of investigation is an experiment.
Can be expressed as a null hypothesis or alternative hypothesis.
Can get directional or non directional hypotheses.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is a null hypothesis?

A

Predicts that the findings / any difference or correlation found is only due to chance, so variables being studied are not causing changes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is an alternative hypothesis?

A

States there is a relationship between two variables being studied and it is IV causing change to DV.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the difference between directional and non directional hypotheses?

A

Directional (one tailed): a specific direction of change or correlation between two variables is predicted, chosen if researchers have an idea of what may happen.

Non directional (two tailed): no specific direction or correlation between two variables is predicted, chosen if researchers are uncertain of what may happen or do not know.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the different types of extraneous variables?

A

Participant variables
Situational variables
Researcher effects
Demand characteristics

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is an extraneous variable?

A

Something that has the potential of affecting the DV that is not the IV.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Describe the levels of measurement used to measure variables.

A

Three types of numerical data used to measure variables:

  • nominal data: totals of named categories
  • ordinal data: data as points on a scale and can be placed in order, not necessarily equal gaps between points
  • interval data: most complex level of measurement, data as points on a scale with equal gals between points e.g. Scientific scales e.g. Cm
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a target population?

A

The group of people psychologists want to be able to generalise their findings to. Sometimes can be as broad as all humans, but others times can be a smaller group e.g. Teenagers, pre school children , or drug users etc.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Describe sampling.

A

Process of selecting participants from the population that are as representative as possible of the target pop.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Describe the four sampling methods used to recruit pps.

A

Opportunity sampling: consists of taking sample from people available at the time or fit criteria you are looking for

Random sampling: where every member of pop has an equal chance of being selected, usually most representative or whole pop as most fair, chooses people at random

Snowball sampling: used if your pop is not easy to contact, involves asking a pp to recommend others who may be appropriate for the study as they may know others similar to them

Self-selected / Volunteer sampling: consists of pps who volunteer for a study when asked in response to an ad etc

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the five common ethical issues with psychological research?

A 
Informed consent
Deception
Protection from harm
Right to withdraw
Privacy and confidentiality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Evaluate the level of measurement, nominal data, that is used to measure variables.

A

Strengths: easy to generate from closed qs, so large amounts of data collected quickly, increasing reliability.

Weaknesses: without a linear scale, pps may be unable to express degrees of response, mode can in,y be used as a measure of spread

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Evaluate the level of measurement, ordinal data, that is used to measure variables.

A

Strengths: more informative than nominal, indicates relative values on a linear scale, easy to generate from likert and rating scales

Weaknesses: gaps between points not equal so a mean cannot be used to assess central tendency, gaps between points only relative so comparisons between pps may be invalid

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Evaluate the level of measurement, interval data, that is used to measure variables.

A

Strengths: easy to generate from closed q’s, scientific measurements are highly reliable and have absolute zero baseline, more informative than nominal or ordinal as points directly comparable

Weaknesses: in interval scales that aren’t scientific measurements, there’s no absolute zero base line for scale so scoring 0 may not mean pps does not demonstrate variable at all, may just be that scale does not measure it.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Evaluate the sampling method opportunity sampling.

A

Strengths: quicker and easier than other methods as pps are readily available.

Weaknesses: non representative as kinds of people available are likely to be limited and similar, making sample biased.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Evaluate the sampling method volunteer / self selected sampling.

A

Strengths: easy because pps come to you and are committed so likely to turn up for repeat testing

Weaknesses: non representative as kinds of people who respond to requests likely to be similar e.g. Have free time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Evaluate the sampling method snowball sampling.

A

Strengths: easy as you only have to find first few pps, a convenient way to find a sample of a particular kind.

Weaknesses: non representative as likely to be similar in ways other than just common characteristics needed for study.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Evaluate the sampling method random sampling.

A

Strengths: should be representative as all types of people in pop are equally likely to be chosen.

Weaknesses: hard to ensure everyone is equally likely to be chosen due to lack of info or access for e.g, and sample may be biased

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Describe primary and secondary data.

A

Primary data: when researchers are working directly with pps through experiments, self reports etc I.e. Collecting data directly from the source

Secondary data: when data is obtained second hand e.g. From studies by other researchers they want to re analyse

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is meta analysis?

A

A type of research methods whereby researchers collect data from several studies to re analyse.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is raw data?

A

Data that psychologists have collected from an investigation bit has not been processed or analysed yet. Usually recorded by being put into a table (tabulating data).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Describe tabulating data in raw data tables.

A
  • often as tally charts of frequency tables e.g. From questionnaires or observations
  • can be reliable as frequencies can be easily checked and used in calculations
  • can draw estimations from the table
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Describe tabulating data in summary tables.

A
  • presenting key findings of a study clearly in a table
  • data can be interpreted in a table
  • labelled clearly and concisely with a title and units, as are all data tables
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

Describe estimation by researchers from their results.

A
  • making a judgement on basis of the data
  • can be helpful when deciding on how to present data
  • should round figures in your calculations to one sf when estimating.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

Describe the different numerical forms in psychology.

A

Percentages: frequently used.

Ratios: Can be used as a comparison between values of different categories, should be ‘reduced’ or simplified where possible.

Fractions: a portion of a whole number.

Decimals: represents a portion of a number, just in a different form.

Standard form: used for v large of v small numbers by showing how many (x10) the no has been multiplied by.

Significant form: simplifying a long figure, commonly either to one, two or three significant figures.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What are the different measures of dispersion?

A
  • the range
  • variance
  • standard deviation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What are the different measures of central tendency in descriptive statistics?

A
  • the mean
  • the median
  • the mode
29
Q

Describe the range.

A
  • difference between largest and smallest value (take one away from other, then sometimes add 1)
  • easy to calculate and takes into account extreme scores
  • but may not reveal full extent of distribution as extreme scores can distort it.
30
Q

Describe the variance.

A
  • tells us more than the range
  • looks at difference between each point (deviation) and the mean and how spread out data is around the mean
  • unlike range, variance takes every score into account so is not distorted by extreme scores
  • but variance is a squared no, so not same units as the mean, (which is why standard deviation is then used)
  • represented by the formula:
Symbols represent:
X= values
X w dash on top= mean
Sideways M= total of all ((X-X w dash on top) squared)s
n= no of values in data

So Formula is:

sideways M (X - X w dash on top)squared ➗ n - 1

31
Q

Describe the standard deviation.

A
  • tell us average distance from the mean of data points, like the variance
  • can tell us how spread out scores are around the mean, the higher the score, the more spread out
  • is representative
  • basically involves everything same as the variance, then just square rooting final answer of variance at the end, done to convert final value to same units as the mean so easier to make a direct judgement.
32
Q

Give brief step by step instructions to calculate the sample variance and standard deviation.

A
  1. Make a table with three columns: X, X-X w dash, and (X - X w dash) squared
  2. Calculate the mean from your scores (X w dash) and use to fill in second column
  3. Then fill in third column
  4. Total up all results in third column to get the sideways M value
  5. Divide that by n-1 to give the variance
  6. Square root variance = answer
33
Q

Describe the mean.

A
  • get by adding up number values and divine by how many numbers there were
  • used with interval data
  • uses all data but one rogue number can badly distort it, so should not be used with extreme values.
34
Q

Describe the median.

A
  • the central number when values are placed in numerical order
  • used with ordinal and interval data
  • not so badly affected by rogue scores but is less sensitive to variations in data and not good for using with small data sets
35
Q

Describe the mode.

A
  • is the most common data value
  • if two values occur most frequently, the data set is then bimodal
  • used with all levels of data including nominal
  • good for revealing how often something occurs, but sometimes there is not one in a data set.
36
Q

What do descriptive statistics include?

A
  • central tendency
  • dispersion
  • graphs
37
Q

Describe the various ways of how data can be displayed.

A 
Bar chart
Pie chart
Histogram
Line graph / frequency polygons
Scatter diagram
Tally chart / frequency tables
38
Q

Describe the types of bar charts

A

Standard:

  • Include numbers of things in diff categories
  • Gaps found between bars
  • Categories on the x axis
  • Frequencies on the y axis
  • used with nominal or ordinal data

Stacked:

  • diff bars relating to same category are ‘stacked’ on top of each other
  • often used when there are two or more conditions or more than one measure of DV

Paired:
- useful if there are two or more levels of IV for a direct comparison

39
Q

Describe pie charts.

A
  • a circular graph divided into slices to illustrate numerical proportion e.g. for percentages
  • data must be expressed as fractions and converted to angles to populate the chart (done by multiplying fraction by 360)
40
Q

Describe histograms.

A
  • like a bar chart, but x axis measures a constantly changing scale (e.g. Mass / height / time), y axis measures frequency
  • a continuous variable on x axis
  • no gaps between bars
  • used with interval or ratio data
41
Q

Describe line graphs / frequency polygons.

A
  • like a histogram, but instead of bars a line is drawn joining all the points
  • a continuous variable on x axis
  • no bars, just a line
  • can show results from two conditions at same time, using two diff lines
  • if a line of best fit is drawn for the points, called a frequency distribution curve
42
Q

Describe scatter diagrams.

A
  • shows values for the same individual but for two diff variables, each plotted on one axis
  • line of best fit drawn
  • used to depict a correlation between two variables
  • can show a positive, negative, zero or curvilinear relationship
  • can also indicate strength of the relationship
43
Q

Describe tally charts / frequency tables

A
  • used for counting things
  • used in content analyses and observations
  • record no of times something is seen
  • category column on left, tally / frequency column on right
44
Q

How is probability used in psychology?

A
  • Inferential statistics allows psychologists to draw conclusions based on the probability that results could have arisen due to chance
  • if findings could not have arisen due to chance, or it is highly unlikely that they have, then we can conclude the findings are significant.
45
Q

Describe the significance level.

A

The probability that a pattern in results has a arisen by chance. Usually set to p

46
Q

Briefly describe what statistical tests do.

A
  • each test calculates an observed value from research data
  • critical value is then elected from a table by comparison to the observed value
  • key to this comparison is the significance level set by the researcher
47
Q

Describe parametric tests.

A
  • more powerful than non parametric tests, but have more stringent requirements for their use
  • all data must come from interval scale and be normally distributed
  • more complicated to calculate
  • do not need to know specific examples of these
48
Q

Describe non parametric tests.

A
  • less powerful than parametric tests and have less stringent requirements for use
  • all data must come from interval scale and be normally distributed
  • specific examples required: spearmans rho, chi-square, Mann Whitney U, Binomial sign, wilcoxon signed ranks
49
Q

What non parametric test do you use if the data and method is…

a) a correlational design with ordinal or interval data
b) an experiment with an independent measures design and nominal data
c) an experiment with an independent measures design and ordinal or interval data
d) an experiment with a repeated measures / matched pairs design and nominal data
e) an experiment with a repeated measures / matched pairs design and ordinal or interval data

A

a) Spearman’s Rho Test
b) Chi-Square Test
c) Mann Whitney U Test
d) Binomial Sign Test
e) Wilcoxon Signed Ranks Test

50
Q

What does the type of non parametric test chosen to use in research depend on?

A
  • the research method employed
  • the experimental design in the case of experiments
  • the level of measurement
  • the distribution of data
51
Q 
What do these symbols mean?
<
><<
>>
~
p
A

: more than
<>: much more than
~: roughly equivalent to, approximately
p: probability

52
Q

Describe the Chi-Square Test.

A

Used when the hypothesis predicts a difference between two conditions.

53
Q

How do you carry out the Chi-Square calculation Test?

A
  • state alternative and null hypotheses (and for alternative specify whether directional or non directional and one tailed or two tailed)
  • draw a contingency table
  • find observed value by comparing observed and expected frequencies in each cell
  • find critical value of chi square using table and using: degrees of freedom (no of (rows-1) x no of (columns-2)), whether test is one of two tailed, the level of significance
  • compare observed val to critical val
  • state conclusion (if observed val is less than critical val, you accept your null and reject your alternative, and results will not be significant)
    Practice
54
Q

Describe Mann Whitney U Calculation Tests.

A
  • used to test for differences between two data sets, so when we have independent groups / measures design
  • data must be ordinal or interval for it to be used
55
Q

How do you carry out the Mann-Whitney U Calculation Test?

A
  • state alternative and null hypothesis
  • record data in a table
  • rank data regardless of group each score is in (lowest score gets rank of ‘1’, next lowest gets ‘2’ etc, if two scores are same, they both get average of the rank they would have obtained)
  • calculate sum of ranks for smaller of two samples to get the observed value (U)
  • find critical value of U (N1 = no of pps in group 1, N2 = no of pps in group 2), also use level of significance and nature of hypothesis for this
  • state conclusion, observed value of U must be equal to or less than critical val in Mann Whitney U for result to be significant.
    Practice
56
Q

Describe Wilcoxon T Calculation Tests.

A
  • used to test differences between two data sets
  • two sets of data must be related e.g. Repeated measures design or matched pairs design for it to be used
  • sets of data must be Oedipal or interval.
57
Q

How do you carry out the Wilcoxon T Calculation Test?

A
  • state alternative and null hypothesis
  • record data in a table
  • calculate difference between scores in two conditions and rank them
  • find observed value of T by totalling ranks for positive differences and then negative differences (ignore ones with no difference), smallest total is observed value of T
  • find critical value of T
  • state conclusion, observed value must be equal to or less than critical value for data to be significant.
    Practice
58
Q

Describe Binomial Sign Calculation Tests.

A
  • used to test for differences between two data sets
  • two data sets must be related e.g. Repeated measures or matched participants design
  • data must be nominal
  • is the least sensitive test compared to others, so may not find a difference that other tests would.
59
Q

How do you carry out Binomial Sign Calculation Tests?

A
  • state alternative and null hypothesis
  • record data in a table
  • cross out scores for any pp that gave same response in both Q’s
  • count up number of pps who remain, this is ‘N’
  • assign a + or - to indicate direction of difference between each pps scores
  • count up no of people given a + and those then given a -
  • observed value of S is the smaller of these two totals, find that
  • find critical value of S in critical values table using significance level, nature of hypothesis and no of pps
  • state conclusion, observed val must be equal to or less than critical val for results to be significant, otherwise accept the null hypothesis.
    Practice
60
Q

Describe Spearman’s Rho Calculation Tests.

A
  • used when hypothesis predicts a correlation between two variables
  • two sets of data must be related e.g. Pairs of scores from one person
  • data must be ordinal or interval
61
Q

How do you carry out Spearman’s Rho Calculation Tests?

A
  • state alternative and null hypothesis
  • record data in a table
  • rank each co-variable from low to high and calculate difference
  • calculate the sum / total of each of the differences squared
  • find observed value for Rho, using correlational coefficient equation, (see notes + learn)
  • find critical value of Rho using table of critical values, no of pps and level of significance
  • state conclusion , observed value has to be equal to or greater than critical value for results to be significant, yet if the sign is in wrong direction, e.g. A positive correlation was predicted / stated but a negative correlation was found, or vice versa, then we have to accept the null hypothesis.
62
Q

Describe a type 1 error.

A
  • refers to a situation in which we have assumed our findings show something, when actually the do not!
  • e.g. Wrongly rejecting a null hypothesis that is true
  • e.g. Wrongly accepting the alternative hypothesis

So a type 1 (One) error is an Optimistic error! Being more optimistic than you should have been.

63
Q

How could the likelihood of type 1 errors occurring be increased?

A

If the p value is set too leniently, because a higher p value increases possibility that results are due to chance

64
Q

Describe a type 2 error.

A
  • refers to a situation in which we may miss something that is actually happening!
  • e.g. wrongly accepting a null hypothesis that is false
  • e.g wrongly rejecting the alternative hypothesis

So a type 2 error is an error where someone wrongly took a ‘better 2 be safe than sorry’ approach, being more pessimistic than they should have been.

65
Q

How could the likelihood of type 2 errors occurring be increased?

A

If the p value is set too stringently as it means we may miss something, yet sometimes this is actually necessary e.g. if you are testing new medical treatments, for example drugs

66
Q

Describe a normal distribution curve.

A
  • in real life situations, normal distribution is usually presented when plotted on graphs, by variables such as height, weight, shoe size, exam results, IQ scores etc
  • have symmetry at mean value
  • curve end points or ‘tails’ meet the x-axis
  • shape of the curve is bell shaped
  • mean, mode and median would all be exactly the same
67
Q

Describe skewed distribution curves.

A
  • when the spread of data scores is greater on one side and mean, median and mode are all in different places
  • can be positively or negatively skewed
  • positive skew = long tail is to the right, measures of central tendency (mean, median and mode also decrease in value in positive skews
  • negative skew = long tail is to the left, measures of central tendency increase in value in negative skews
68
Q

How are research reports organised when writing them?

A

-

PSYCHOLOGY AS (17 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions