Research Methods Flashcards
1
Q
Informed consent
A
Ppts should be told as much as they can be about the study so they can make a decision as to whether or not to partake.
In some cases, however, PPts cannot make such decisions because they may not be able to undertsand. In such cases, the consent of the parent or guardian should be taken. Or the ethics committee should be contacted.
Some issues with this is that the researchers may not know what will happen and ppts may not show natural behavior.
2
Q
Deception
A
Misleading the ppts or withholding the true aim of the study because the ppts are likely to object or show unease once debriefed.
It should only be used as the last alternative, approval should be gotten from the ethics committee, there must be a debrief and prior general and presumptive consent should be gotten.
Deception may ensure natural behavior and avoid the Hawthorne effect. Effective debriefing may reduce harm caused by deception.
3
Q
Protection from harm
A
Ppts may not be exposed to great psychological or physical rish than their normal life experiences.
There must be a debrief with true aim of the the study, right to withhold data, reassurance of normal behavior and offering counsel
Sometimes harm is necessary
4
Q
Right to withdraw
A
Ppts must be made aware that they are allowed to leave at any time and refuse permission to have their data used.
Ppts must be clearly informed of this and procedures should be put in place for this to happen.
It may lead to incomplete experiments
5
Q
Confidentiality
A
Information about ppts is protected by data protection act and they must be identifiable in research.
Ppts are given numbers or referred to by initials.
Initials may not always be confidential.
6
Q
Privacy
A
This involves not invading people’s personal lives which may be difficult if ppts are unaware they are being watched.
Ppts should only be observed in public.
Some very important issues only take place in private. Hawthorne effect.
7
Q
Prior general consent
A
Obtaining prior consent from a ppt to see if they would be okay partaking in a study involving deception.
8
Q
Presumptive consent
A
Taking a random sample of the population and introducing them to the research including deception and if they agree, this can be generalised to the general population.
9
Q
Retrospective consent
A
Once the true nature of the study has been revealed, the ppt has the right to withdraw
10
Q
Variable
A
Any factor, trait or condition that can exist in differing amounts or types
11
Q
Independent Variable
A
One that is changed by the researcher
12
Q
Dependent Variable
A
One that is affected by the independent variable
13
Q
Extraneous Variable
A
Potential variable that can affect the invetsigation
14
Q
Laboratory Experiments
A
They look for the effect IVs have on DVs in a controlled, artificial environment where the participants are allocated randomly to experimental conditions
15
Q
Strengths of lab experiments
A
High control over extraneous variable so causation can be asserted
Reliable results as created conditions can be replicated
Variables can be measured accurately and empirically
16
Q
Weaknesses of lab experiments
A
Lacks ecological validity because it is artificial
High risk of demand characteristics
Experimental bias; how researchers interact with ppts
17
Q
Field Experiments
A
They are conductded in a natural setting but the IV is still altered by the researcher
18
Q
Strengths of field experiments
A
Higher ecological validity
Demand characteristics are less of an issue
19
Q
Weaknesses of field experiments
A
More extraneous variables
Lack of informed consent
Poor reliability
Sample bias because ppts are not randomly allocated
20
Q
Natural Experiments
A
When researchers investigate a naturally existing change as their IV
21
Q
Strengths of natural experiments
A
High ecologoical validity
Less demand characteristics
Can be used in situations where manipulating the IV would be unethical
22
Q
Weaknesses of natural experiments
A
Sample bias
Extraneous variable reduces causal effects
Lack of informed consent
23
Q
Quasi experiments
A
IV is alredy existing
24
Q
Strengths of quasi experiments
A
They can be carried out under controlled conditions
Allows areas of research otherwise impossible
25
Weakness of quasi experiments
No control over assignment of ppts to independent variable
26
Questionnaires and its different formats
They are written self-report technique where ppts have a set number of answers to respond to.
They can be with closed (where there is a predetermined set of data and which produces quantitative data which may lack detail but is easy to analyse) or open questions (where there is no restriction on how ppts can reply which is rich in detail but hard to analyse)
27
Types of questionnaires
Likert scale: indicate agreement on a scale of 5 points
Rating scale: identify a value that represents their strength feeling about a particular topics
Fixed choice option: list of possible options and respondents are required to indicate what applies to them
28
Strengths of questionnaires
They are relatively cheap and quick to gather large amounts of data
Ppts are more likely to be honest because they are anonymous
29
Weaknesses of questionnaires
Social desirability bias
Sample bias may be an issue as only certain types of ppl answer them
Some questions may be leading
Response bias; e.g acquiescence bias due to an overuse of jargon or double-barrelled questions which either use double negatives or 2-in-1 questions
Questions may be misunderstood
30
Interviews and its different types
They are self-report techniques that involve the researcher asking questions and recording the answers either on a one-to-one basis or with multiple ppts are the same time.
Structured: list of questions rigidly stuck to, known as interview schedule which allow for easy replication and generalised findings.
Unsturctued: list of topics of questions with extra flexibility which gives more detailed discussion which does not allow for replication or generalisation
31
Strengths of interviews
They gather more rich and detailed information than questionnaires
They are best suited for complex or sensitive issues
They can be used as part of a prior study to gather information prior research
32
Weaknesses of interviews
A lot of time and expenses goes into training interviewers
Social desirability bias
Detailed data may be difficult to analyse
They require ppts to have basic competencies such as memory which could reduce the sample size.
33
Types of observations
Controlled, Naturalistic, Participant and Non-Participant Observations, and Covert and Overt Observations
34
Controlled Observations
They are likely to be carried out in an artificially controlled environment, with a standardised procedure and a behaviour schedule (agreed scale of coding behavior) and observing behavior in categories e.g Strange Situation
35
Strengths of controlled Observations
Easily replicated so easy to test for reliability
Quantitative to easy and quick to analyse so less time-consuming
Fairly easy to conduct so a large sample can be obtained to be generalised to a large population
36
Weakness of Controlled Observations
Lack validity due to demand characteristics and ecological validity
37
Naturalistic Observation
Studying natural behaviour of ppts in natural surroundings which could use behavioural categories or not e.g William's study of aggression after intro to tv
38
Strength of a naturalistic observation
Greater ecological validity
39
Weaknesses of naturalistic observation
Micro scale so not representative so not generalisable
Less reliable so difficult to replicate
Need to be trained to recognise psychologically significant
Less able to manipulate variables so causation uncertain
40
Participant observation
One where the researcher joins the group and observes their activities by writing notes and reflecting later on e.g Rosenhan reliability of diagnosis of schizo
41
Strength of participant observation
Close proximity means unlikely to overlook behaviour--increased insight
42
Weakness of participant observation
Hawthorne or investigator effect
43
Non-participant observation
The psychologist observed activities but does not take part in them e.g Strange Situation
44
Strength of non-participant observation
Ppt reactivity and investigator effects less likely
45
Weakness of non-participant observation
Rsr might overlook or miss interesting behaviour
46
Covert and overt observation
Covert is when the rsr real identity and purpose is kept concealed
Overt is when the rsr revels their true identity and purpose and asks for permission to observe
47
Strength and weakness of covert observation
Increased validity due to increase of natural behaviur
Ethical Issues due to lack of consent
48
Strength and weakness of overt observation
Ethically acceptable
Low internal validity due to influenced behaviour
49
Structured and unstructured observation
Structured clearly defines and focuses on behaviours linked to study
Unstructured writes all that's seen
50
Strength and weaknesses of unstructured observation
Rich qualitative data which increases internal validity
Time consuming to complete, analyse and train researchers
Risk of observer bias reducing reliability
51
Strengths and weakness of structured observation
Time efficient due to qualitative data
Easier to replicate because its reliable
It may lack important details
52
Behavioural categories
When target behaviour is broken up into components that are observable and measureable
53
Strength of behavioural categories
The result is reliable and quantitative and can be used to identify trends and patterns
54
Weaknesses of behavioural categories
Researchers may interpret information differently leading to unreliable results
Categories need to be valid
Significant behaviour may occur that is missing from categories
Time is required for observers to train and practise
55
Types and explanation of sampling methods
Event sampling: the number of times specific event or behavior occurs during the observation
Time sampling: any behaviour being demonstrated at a specific time interval
56
Strength and weakness of event sampling
Behavior could occur infrequently and could be missed by time sampling
It might miss other important complex behaviours
57
Strength and weakness of time sampling
Useful in reducing the amount of observations needed to be taken
Might miss infrequent behaviours
58
Inter-observer reliability
The idea that two researchers must be consistent and record similar data for unbiased and objective results.
59
Necessities for Inter-observer reliability
They must:
- familiarise themselves with the categories
- observe the same thing simultaneously in a pilot study
- analyse the collect data by correlating each pair of results
60
Correlation
A type of non-experimental research that measures the strength and direction of a relationship between co-variables
61
Types of correlations
Positive: change in the same way
Negative: change in different ways but there is a pattern
No correlation: no pattern
62
Strengths of correlations
They are useful as a preliminary research technique to identify a link before further research
Can be used to research unethical topics
Provides a precise and quantifiable measure of how two variables are related
Quick and easy to carry out
63
Limitations of correlations
Explains why variables are related but not why
It does not establish causation
Does not take into account third-factor variable which impacts both variables so they don't affect each other
64
Correlation coefficient
The strength of a correlation which is between 1 and -1
65
What is a strong, moderate and weak correlation coefficient?
Strong: 0.7 or above
Moderate: 0.4 or above
Weak: 0.1 or above
66
Content Analysis
Systematically reviewing qualitative information to produce a conclusion and test a hypothesis which can be done by coding or thematic analysis
67
Strengths of a content analysis
Data is often easily available
Cheap and easy
Well-designed coding units means valid and easily reliable
Few ethical issues
Real-life sources give high external validity
Can produce quantitative data which is easy to analyse
Qualitative thematic analysis which is rich in detail and meaning
68
Weaknesses of content analysis
Thematic analysis can be time consuming
Element of subjectivity required to interpret data
It could show possible experimenter bias and reduce reliability and internal validity
It does not give any causal explanation
69
Coding analysis
Qualitative data is place into categories based on coding units, how many times a word or phrase appears in a text
70
Thematic analysis
Wider themes or idea are present and used to make decisions
71
Case studies
A very detailed investigation of an individual or small group of people usually regarding an unusual phenomenon or biographical event of interest to a research field
72
How are ppts interviewed in case studies
Interviews, family history, questionnaires, autopsies, cognitive tests, school records, observations, experiments and longitudinal studies
73
Strengths of case studies
Rich yield of data
Can give insight into how something functions
Longitudinal study can study change over time
It could conflict with current theories and stimulate new, better paths for research
74
Limitations of case studies
Small sample size cannot be generalised
Little control over variables
Poor reliability as replication is unlikely
Researcher may become so involved that they exhibit bias which reducing how factual it is
75
Reliability
The consistency of findings of research which often depends on how replicable it is
76
Internal reliability and how it can be checked
The consistency of results across items within a test
It can be checked by split-half method; items or questions and data collected is split randomly in half and compared to see if results taken from each part of the measure is similar
77
External reliability
The extent to which a measure varies from one use to another e.g experiment carried out on two different days
78
How to assess external reliability
Inter-rater reliability: the degree to which different raters give consistent estimates of the same behaviour
Test-retest reliability: used to assess the stability of a test over time by giving ppts the same test on different occasions
79
Ways to improve reliability (self-report)
Rewrite questions that give inconsistent data
Structured approach as they are more specific
Use the same interviewer to reduce bias
Trained interviewers
80
Ways to improve reliability (experiments)
Operationalised DV and IV
Standardised procedure
Controlled extraneous variable
81
Ways to improve reliability (observations )
Train observers in observational techniques
Operationalised behavioural categories
82
Validity
Refers to whether a test actually measures what is claims or whether a theory really explains the phenomenon
83
Internal Validity
Whether the effects observed in the study are due to the manipulation of the IV and not other factors
84
External Validity
The extent to which findings can be generalised beyond research findings
85
Population validity
How far the results can be generalised to other groups
86
Ecological validity
To what extent the results can be generalised to real-life settings
87
Temporal validity
The extent to which results can be generalised to other time periods
88
State and explain how to assess internal validity
Face Validity: looking at the test at face value and using intuition to see whether the test is appropriate for measuring phenomenon
Concurrent Validity: a measure of how well a particular test correlates with previous validated measure; having a high concurrent validity of 0.8
89
State how to assess external validity
Population, ecological and temporal validity
90
How to improve validity (experimental design)
Using a control group
Standardised procedure
Blind/double blind
Clear IV or DV (operationalised)
Reduce EV or nuisance variable
Keep true aim covert to reduce demand characteristics
Use of matched pairs or repeated measures to reduce ppt variables
91
How to improve validity (questionnaires)
Lie scale (varied comments and contradictory statements to check consistency of results)
Anonymity reduces demand characteristics or social desirability bias
92
How to improve validity (observations)
Covert observations
Behavioral categories
Naturalistic
93
How to improve (qualitative methods)
Triangulation: use of self-report techniques and observations before reaching a conclusion
Interpretive validity: reporting or use of direct quotes
94
Experimental design types
Independent groups, matched pairs and repeated measures
95
Independent groups
Different ppts are randomly allocated to each conditions of the independent variable ensuring they have an equal chance and reducing investigator bias
96
Order effects
When a ppt becomes tired, bored or fed of with repeating the same conditions or they begin to guess the true aim of the experiment
97
Advantage of independent group
Reduces order effects as they are in one condition only
98
Disadvantages of independent group
Ppt variables, such as variations in age or sex, may affect results
More ppl are needed than in other designs
99
Repeated measures
The same groups of ppts take part in each condition of the independent variable
100
Advantages of repeated measures
Fewer people are needed which saves time
Reduces effect of ppt variables
101
Disadvantages of repeated measures
Order effects
Demand Characteristics
102
Counterbalancing
Alternating the order in which ppts perform the different conditions of the experiment
103
Matched pairs
Ppts are matched on key characteristics and one member is allocated to one test condition
104
Advantages of matched pairs
Reduces ppt variable as ppts are similar
Avoids order effects so counterbalancing is not necessary
105
Disadvantages of matched pairs
Very time consuming to find closely matched pairs
Impossible to match people up exactly
106
Aim
Identifies the purpose of the investigation
107
Hypothesis
A precise, testable statement of what the outcome of the study could be, proposing the relationship between two operationalised variables
108
Types of hypotheses
Null hypothesis: assumes no relationship between IV and DV
Alternative hypotheses: assumes a relationship either a specific (Directional or one-tailed) or not specific (Non-directional or two-tailed)
109
Pilot study
Small-scale trial that runs off the actual investigation using a smaller sample size
110
How to use pilot studies to improve self-report methods or observational studies
Self-report: try out questions in advance and remove ambiguous or confusing ones
Observation: check behavioural categories
111
Advantages of pilot study
Identify potential issues in procedure that saves money and time in the long run
Removes extraneous variables and increases internal validity
Maintaining control so the study is more reliable
112
Sample
A group of ppl who take part in an investigation
113
Sampling
The process of selecting ppts from the target population
114
Target population
The total group of individuals from which the sample might be drawn
115
Generalisability
The extent to which findings from the research can be applied to the target population
116
Purpose of sampling
Impossible to study every single individual so sample is representive which is more realistic
117
Random sampling
Each member of the target population has a chance of being selected
118
Sample frame
Data on target population to select from
119
Advantage of random sampling
No researcher bias, more representative population
120
Disadvantages of random sampling
Can be time consuming e.g sourcing different lists
Some may refuse; volunteer bias
121
Systematic sampling
Ppts are selected at fixed intervals from the sample frame e.g every fifth ppt
122
Advantage of systematic sampling
No researcher bias, more representative
123
Disadvantages of systematic sampling
Can be time consuming
Might lead to volunteer bias
124
Stratified sampling
Reflects main groups of target population
125
Strata
Subgroup of target population
126
Advantages of stratified sampling
Generalisable if strata is accurate
Removes researcher bias
127
Disadvantage of stratified sampling
Time consuming due to requirement of accurate information
128
Opportunity sampling
Researcher selects whoever is available at the time
129
Advantage of opportunity sampling
Quick and easy way of selecting information
130
Disadvantage of opportunity sampling
May not be representative and some research bias
131
Volunteer sampling
Ppts put themselves forward
132
Advantage of volunteer sampling
Easy, quick and cheap
133
Disadvantage of volunteer sampling
Volunteer bias, not representative which reduces population validity
134
What are the two main variables in an experiment?
Independent variable (the one being manipulated by the researcher) and the dependent variable (the one affected by the IV)
135
What are the unwanted variables
Extraneous or nuisance variables: variables that are not the IV that affect the DV
136
Types of extraneous variables
Participant variables: features of ppts
Situational variables: features of experiment or setting e.g temperature
Demand charateristics: clues that imply aim
Experimenter bias: any intentional or unintentional influence of the investigator's behavior which affects the investigation
137
Confounding variables
Change systematically with the IV and affects the DV
138
Randomisation
Making elements of the study unpredictable to reduce any extraneous variables e.g random words for recall or random conditions
139
Standardisation
Using exactly the same procedure and instructions for all ppts in a research study to reduce extraneous variables and increase replication
140
Single blind vs double blind
Researcher knows IV groups but ppt doesn't
Neither researcher nor IV knows IV groups
141
Qualitative data
Expressed in written description of thoughts and feelings from accounts of interviews or diaries
142
Strengths of qualitative data
Rich in detail
Higher in validity as it provides more meaningful insight of ppts view
143
Weaknesses of qualitative data
Can be difficult to analyse data and occuring patterns
Conclusions could be subjective
144
Quantitative data
Expressed numerically and can be analysed statistically
145
Strengths of quantitative data
Simple to analyse
Easy comparison between groups
Less biased numerical data
146
Weaknesses of quantitative data
May fail to represent real life
May not include thoughts and feelings
147
Primary data
Collected for the purpose of the investigation by the researcher
148
Strength and weakness of primary data
Specifically designed for experiment
Time-consuming
149
Secondary data
Data that already exists collected for the research
150
Strengths of secondary data
Easy and cheap
No need for additional collection
151
Weaknesses of secondary data
Outdated
Incomplete
May not match needs of the investigation
152
Advantage and disadvantages of mean
Most sensitive measure of central tendency as it is made up of all scores in data set
Anomalous data could distort it
May not make sense using discrete units
153
Advantage and disadvantages of median
Less affected by extreme scores than the mean
Less sensitive than the mean as it does not reflect all data
Not suited for smaller data sets
154
Advantages and disadvantage of mode
Only way of describing data from categories
Normally unaffected by extreme scores
May not be central measure as it does not represent all data
155
Range
The difference between the highest and lowest scores in a data set, with the addition of 1
156
Advantage of range
Provides direct information on the data which is easy to analyse
157
Disadvantages of range
Affected by extreme values
Does not take into account all the data
Doesn't consider the distribution of data within range
158
Standard variation
How much variation from the average exists
159
Advantage and disadvantage of standard variation
More precise measure which takes all values into account
Can be affected by anomalies if there are a few (not just one)
160
Measures of Central tendency
Mean, median and mode
161
Measures of dispersion
Range and Standard Deviation
162
Discreet Data
Exact figures which can be counted and can be put into categories with no relationship
163
Continuous data
Measurable values representing a range of information on a scale with relationship between data
164
Types of graphs
Bar graphs, histograms, line graphs, scattergrams and tally charts
165
Bar chart
Used for discreet data, where the IV is plotted on the x axis and the DV on the y axis and with a title where both are operationalised
166
Histograms
Used for continuous data and the bars can be joined
167
Line graph
Used for continuous data with a curve of best fit, showing change over time on the x axis
168
Scattergrams
For correlations with a line of best fit
169
Distribution Curve
One used to represent frequency data (on the x axis) in a population or sample
170
Normal distribution
A classic bell-shaped curve with the mean, median and mode in the middle and about 68% falling with the midpoint, then 95% in the whole curve as the tails never touch the x axis for more extreme figures
171
Positive distribution
The tail is to the positive side with the order of value being: mode, median and mode; here most people get low scores
172
Negative distribution
The tail is to the negative side with the order of values being mean, median and mode where most people get high scores
173
Peer review
Takes place before a study is published to ensure that it is of high quality by subjecting it to scrutinisation by a small group of experts in a particular field unknown to the author or researcher
174
Aims of peer review
To allocate research funding
To validate the quality and relevance of research
To suggests amendments and improvements
175
Evaluative points of peer review
Anonymous peer reviews; give overly critical reviews to protect prestige; naming reviewers to avoid this; valid research may be blocked---Anonymity could lead to more honest opinions
Publication Bias; more likely to publish one with results than null results; file drawer effect; important findings not published and limitation of possible hypotheses
Buried groundbreaking research; expert reviewers are invested in theories, suppress opposition to mainstream theories; new theories suppressed; important findings may not be published
176
The Economy
The consumption and production of goods within a country measured as a currency value
177
The implication of psychological findings for the economy
What we can learn from research that can influence, benefit or devalue our economic prospoerity
178
Impacts of economic implications
Development of treatments for mental illnesses (reduced absences at work), e.g systematic desensitisation
Further research into important topics such as role of the father (both parents can provide emotional support, more flexibile working arrangements)
Improving testimonies in court e.g accurate EWT reduces expenses on wrongful arrests
Drugs for dementia saves money than therapy
Social changes such as campaigning to reduce drunk driving and smoking
179
What are the features of science
Empirical evidence
Objectivity
Replicability
Theory Construction
Hypothesis testing
Falsifiability
Scientific Paradigms
180
Empirical methods
Using direct observations or experiments rather than unfounded beluefs to look for facts
181
Objectivity
Empirical data should not be affected by expectations or opinions of the researcher as they are conducted in controlled conditions and measured concepts are operationalised
182
Replicability
To what extent an experiment can be repeated to get similar results by different groups of people to see if similar behaviour is observed, ensuring the theory is generalisable
183
Theory construction
Explanations must be constructed to make sense of the facts to predict behaviours or natural phenomena using inductive or deductive reasoning
184
Inductive Reasoning
A theory is formed from an observation where a pattern is noted
185
Deductive Reasoning
A theory is formed first then an hypothesis and then an observation is conducted to confirm it
186
Hypothesis testing
Theories must be able to generate testable expectations or it requires modification
187
Falsifiability
The idea that a theory must be able to be proved false by observation or experimentation
188
Paradigm
Unifying theory which may shift when a new theory is found commonly accepted by member of a discipline or group
189
Three Stages of scientific discipline
Prescience: range of views and no general paradigm
Normal science: generally accepted paradigm and research within discipline
Revolutionary science: paradigm shift; assumption and ways of studying language change
190
Paradigm shift
Describes a change in view and new way of investigation
191
Scientific Paradigm
An idea by Thomas Kuhn that all science must have a unifying theory, but social sciences, such as psychology, do not
192
Features of reporting a psychological investigation
Title: clear and precise statement on investigation
Abstract: short summary of study including APFC which appears first but is written last
Introduction: review of theories and reason for study and funnel technique which started off broad but is later narrowed down
Methods: should be clear and detailed for replication including ppts, design, apparatus and procedure
Results: descriptive statistics (summarised data in graphs or tables) and inferential statistics (whether or not research is significant and hypothesis is accepted)
Discussion: summary of findings, real-life applications and practical and theoretical implications
References
Consent form
193
Formats of references
Name of researcher, Year, Study, Journal Title, Volume Numbers and Page Numbers
Author, Date, Book Title, Place of Publication and Publisher
194
Format of consent form
Dear ppt,
require ppt agreement
space to sign, date and name
no pressure to consent, right to withdrawal, confidential and anonymous
experiment details
195
Probability
The likelihood an event will happen, expressed as a number between 0 and 1
196
Statistical Significance
Discover if results are significant; not likely to have occurred through change using levels of 10%, 5% or 1% (or 0.1% in stringent conditions)
197
What happens when a result is significant or insignificant?
Significant: reject the null hypothesis
Insignificant: accept the null hypothesis
198
What level of significance is often used
5% or 0.05; 1 in 20 results could have occurred due to chance
199
Nominal data
Data that is mutually exclusive and with no numerical significance
200
Ordinal data
Data in which the order is significant but there is no clear difference between each one, typically meausring non-numerical concepts
201
Interval data
Numerical scales in which we know the order and the exact difference between values
202
Test of Difference vs Test of Correlation
Difference: investigation that measures the difference between two conditions after manipulation of the IV
Correlation: investigation into the relationship between two co-variables
203
Unrelated design
Independent groups
204
Related design
Repeated measures and matched pairs
205
Statistical tests
Chi-squared, Sign Test, Chi-squared,
Mann Whitney, Wilcoxon, Spearman's rho,
Unrelated t-test, Related t-test, Pearson's r
206
Parametric meaning and tests
Level of measurement if interval and the measure of central tendency is mean.
Unrelated t-test, Related t-test and Pearson's r
207
Non-parametric tests
Levels of measurement are ordinal or nominal and median and mode are used as the measures of central tendency.
Chi-squared, Sign test, Mann-Whitney, WIlcoxon, Spearman's rho
208
How to find whether or not a set of data is significant using the sign test
Find the calculated value and n number and use both to find the critical value. If critical is more than or equal to calc value, it is significant
209
The rule of greater r
Any statistical test with the letter r in the name is where the calc value needs to be greater than or equal to the critical value
210
Type 1 error
A situation when an assumption is when that findings show something that they don't; rejecting the true null hypothesis and accepting wrongly the alternative hypothesis both due to chance
211
Causes of type 1 error
P-value too high or lenient, increasing the possibility that results were due to chance Having smaller sample sizes
The probability of making a type 1 error = alpha; using lower level for alpha (level of significance; 0.05) = less likely to detect a true difference is one exists
212
Type 2 error
Missing something that is actually happening; falsing accepting the null hypothesis and rejecting the alternative hypothesis by assuming the results were due to chance
213
Causes of type 2 error
P-value is too stringent (0.01) even in necessary situations such as testing new drugs
The probability of making a type 2 error = beta; ensure test has enough power by having a larger sample since to detect a pract difference when it exists.
