Quantitative Research Flashcards

1
Q

Week 1
Identify the contributions of Descartes and Locke to understanding the mind and behaviour

A

Descartes
- Rationalism - use of reason and logic to derive the truth; senses can deceive
- Cartesian dualism - mind-body dualism - mind and body are conceptually separate

Locke
- Disagreed with Descartes
- Empiricism - knowledge is constructed by experiences and sensation
- Theory that environment shapes the mind was later adopted by Behaviourism movement

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

Week 1
Recognise the work of Fechner and Wundt on the development of experimental psychology

A

Fechner
- Developed psychophysics
– Relations between sensations and the physical stimuli producing them
- First experimental psychologist
- United the mind and body mathematically
- Led to developments of psychometrics and experimental psychology

Wundt
- Established the first psychology laboratory at the University of Leipzig in 1879
- “Physiological psychology”
- Structuralism - breaking down mental processes into their basic elements
- Use of introspection: training people to objectively analyse the content of their own thoughts
- Developed the first journal of Experimental Psychology. ‘Philosophical Studies’ (1883)

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

Week 1
Describe the contributions of Darwin and Galton to methods and analysis used in psychology

A

Darwin
- Places ‘human nature’ in the wider context of evolutionary change
- Functionalism - how human behaviour and mental processes serve to adapt an individual to an ever-changing environment
- Led to the systematic study of individual differences: intelligence and personality tests
- First scientific attempt to study emotions

Galton
- Intelligence is inherent - nature plays no role
- Argued for eugenics
- Measured human ability and classified it - individual differences in human abilities
- Questionnaire design
- Normal distribution
- Intelligence tests (& eugenics)
- Correlation
- Twin study methodology

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

Week 1
Give examples of inductive and deductive reasoning

A

Inductive
- Reasoning from singular statement (or premises) to the probable validity of a conclusion
- Observation -> Pattern -> Hypothesis -> Theory
- Dog has always barked at postman -> postman arrives -> dog will probably bark

Deductive
- Reasoning from general statements (or premises) to a logical and certain conclusion
- Theory -> Hypothesis -> Observation -> Confirmation
- Only dogs bark -> subject barks -> subject is a dog

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

Week 2
Appreciate why we conduct research in psychology, and the role of evidence in challenging common-sense beliefs and intuition

A

Why We Conduct Research
- Process of generating and testing new ideas
- The world is changing - research is needed to further our knowledge
- Avoid myths from intuition
- Studies and theories are underpinned by research

Role of Evidence
- To inform knowledge
- Challenge myths - e.g., Does mobile phone use negatively impact on children’s grammar? (Wood, Kemp & Waldron, 2014)
- Psychology would not exist without research

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

Week 2
Appreciate the importance of using hypotheses in quantitative psychology

A
  • Popper (1963) - theories should be phrased in a way that makes it possible to show how they could be wrong
  • Must be testable, precise and falsifiable
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Week 2
Be aware of the stages in the hypothetico-deductive method

A
  1. Identify a problem
  2. Define a problem
  3. Generate hypothesis/es to test the problem
  4. Design the research to test hypothesis/es
  5. Collect the data
  6. Analyse the data
  7. Interpretation of the results
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Week 2
Explain why generalisability and replicability are fundamental to quantitative research in psychology

A

Generalisability
- Only a small (but representative) sample is needed to generalise to a wider population
- Statistics can be used to generalise

Replicability
- To ensure that the generalisations we make from our research samples hold true for the population
- Theories are not based on only one study
- Avoid ‘flukes’
- Increase confidence in our results
- Can involve minor modifications

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Week 2
Recognise the methods of data collection used in quantitative psychology, namely experiments (including quasi experiments and clinical trials/RCTs), correlational studies, and questionnaires

A

Experiments
- Randomised Controlled Trials (RTCs)
– Most rigorous form of research
– Used to measure the effect of an intervention by ranomly assigning individuals to intervention or control group
–Participants and researcher are blinded to the condition
- True Experiments
– Laboratory-based and fully controlled
– Experimental manipulation
– Standardised procedure
– Random allocations of participants to conditions
- Quasi Experiments
– Like a true experiment, but lacks either/both:
– Random assignment
– Full control over the independent variable

Correlational Studies
- Used to determine a relationship between factors
- Non-manipulated variables
- Useful when you are unable to perform an experiment
- Observe natural variation in variables and measure the correlation between both

Questionnaires
- Commonly used to collect data in correlational studies
- Used to objectively measure a particular concept
- Psychometrics: area of study concerned with the theory and technique of psychological measurement
– Diagnosis or screening for clinical purposes
– E.g., Beck Depression Inventory, Eating Disorders Examination Questionnaire

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

Week 3
Be able to identify the independent and dependent variables of a study

A

Independent
- Variable that the experimenter manipulates as a bases for making predictions about DV

Dependent
- Variable that is measured or recorded in an experiment (outcome)

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

Week 3
Appreciate the different types of variables used in psychology (continuous, discrete, categorical) and be able to identify variables in each category

A

Continuous
- Can take value within any given range, doesn’t change in discrete jumps
- Temperature, anxiety levels

Discrete
- Can only take on certain discrete values within the range
- Number of cars owned, number of children in a family

Categorical
- Value that the variable takes is in a category
- Gender, occupation, ethnicity

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

Week 3
Recognise the differences between within- and between-subjects designs and when it is suitable to use each, and appreciate the use of counterbalancing

A

Within-Subjects
- Repeated measures
- Same participants in every condition of the IV
- Less participants needed

Between-Subjects
- Independent groups
- Different participants in each condition of the IV
- Lots of participants needed

Counterbalancing
- One half of participants complete Condition A first; one half complete Condition B first
- Spreads order effects across both conditions of the IV
- Full or partial

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Week 3
Appreciate what ‘sample’ and ‘population’ mean in psychology research, and the different types of sampling methods most commonly used in quantitative research

A

Sample
- The group selected from the population to participate in research

Methods
- Probability-based - everyone in the target population has an equal probability of being selected
– Simple random - Every member of the population of interest has an equal chance of being selected
– Systematic random - select every nth from the population
– Stratified random - random sample from various sub-sections of the population
- Non-probability - sample is not structured to approximate the population
– Opportunity/convenience
– Self-selected
– Online

Population
- A group that shares a common set of characteristics; the wider group to learn about

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

Week 4
Appreciate what is meant by the term ‘ethics’, and why it is important that ethical guidelines are adhered to in research and applied psychology

A
  • Responsible and morally right conduct
  • Psychologists have a duty of care to protect human and animal participants and clients from harm
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Week 4
Be familiar with the BPS Code of Ethics and Conduct (2021) and the BPS Code of Human Research Ethics (2021)

A

BPS Code of Ethics and Conduct (2021)
- Respect - dignity, privacy and confidentiality, informed consent, no inappropriate use of power, compassion
- Competence - provide services to a professional standard
- Responsibility - professional accountability, use knowledge and skills appropriately
- Integrity - honest, unbiased and fair, avoid conflicts of interest and maintain personal and professional boundaries

BPS Code of Human Research Ethics (2021)
- Risk
- Valid consent & right to withdraw
- Confidentiality
- Deception
- Debriefing

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

Week 4
Be able to describe the following issues related to ethics: risk; consent; the right to withdraw; confidentiality; deception; and debriefing

A

Risk - Any potential physical/psychological harm, discomfort or stress to human participants that a research project may generate

Consent - Agreement to freely and voluntarily participate with the full knowledge of the content and participant rights
Valid - Over 18’s, Under 16’s with parental consent

Right to withdraw - Participants should be made aware of the voluntary nature of research and their right to withdraw at any time with no adverse consequences or penalty

Confidentiality - Information should be appropriately deidentified and not be able to be traced back to them

Deception - Deliberately providing incorrect information

Debriefing - Informing participants about the full nature or rationale of a study after participation and attempting to reverse any potential negative side effects

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

Week 4
Be able to describe ethical issues posed by contemporary areas including: internet research, clinical trials, and research using animals

A

Internet Mediated Research
- Valid consent - ensuring age of participants, ensuring that participants have fully engaged in consent procedures
- Confidentiality - IP addresses - issues with anonymity
- Debriefing - those who don’t participate for the full duration may not be debriefed

Clinical Trials
- Risks - do benefits outweigh risk?
- Is informed consent truly possible?
- Placebo and deception

Research with Animals
- Does the research threaten health and wellbeing?
- Is it fair to study animals to improve the human condition?

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

Week 4
Begin to develop a personal awareness of how to act ethically as a psychologist

A

Acting Ethically
- Collective responsibility - ask if unsure

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

Week 5
Be aware of the different levels of data measurement (nominal, ordinal, interval and ratio) and be able to identify examples of each type of data

A

Nominal
- Categorical
- Gender, ethnicity, job type
- Numbers are given to distinguish between categories, with no particular order to rank importance
- 0 male, 1 female, 2 non-binary, 3 other, 4 prefer to not say

Ordinal
- Categorical
- Using a scale to put people into an order/rank
- E.g., position in a race: 1st, 2nd, 3rd…
- Size of number does represent something
- However, size or difference between numbers, nor the ratio, is informative

Interval
- Quantifiable
- Puts scores in order, however differences between numbers are equal
- E.g., temperature, 0-10 = 10-20
- However, 10oc isn’t half as warm as 20oc
- There is no absolute 0 - 0oc doesn’t equate to 0, or no heat/temperature

Ratio
- Quantifiable
- Same as interval data, but there is an absolute zero
- E.g., height, test score, speed of a car

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

Week 5
Distinguish between different types of the central tendency, namely mean, median and mode, and recognise when it is appropriate to report each

A

Mean
- Sum of scores divided by number of scores in the sample
- Most commonly reported
- Most appropriate for ‘normal’ data

Median
- The middle score/value once all scores in the sample have been put in rank order
- Less commonly reported than the mean
- Non-normal data

Mode
- Most frequently occurring score/category of scores
- Least commonly reported, useful for categorical variables
- Bimodal = 2 modes; Multimodal = several modes

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

Week 5
Explain what population mean and sample mean are, and recognise the issue of sampling error

A

Population Mean
- Typical score in a population
- E.g., population mean for IQ = 104

Sample Mean
- The mean score of the sample taken from the population

Sampling Error
- The difference between the sample statistic and the population statistic
- Generally, the larger the sample, the closer the sample and population means will be

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

Week 5
Be familiar with bar charts and histograms and their uses

A

Bar Charts
- Used to summarise a categorical variable
- x axis represents the categorical variable
- y axis represents frequency, average, percentage, etc.
- Separate bars - unrelated categories

Histograms
- Type of bar chart for continuous variables
- Bars not separated & equal width
- Illustrate whole data set
- All values are represented, even if empty
- x axis: details of score on variable
- y axis: frequency of scores’ occurrence

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

Week 5
Identify methods to illustrate variation in your data, namely: range, interquartile range, and standard deviation

A

Range
- Distance between lowest and highest score in a sample
- Sensitive to outliers (unusually extreme scores)

Interquartile Range
- Distance between the upper (third) and lowest (first) quartile in a set of data
- Appropriate for ordinal level data
- Appropriate for non-normal data
- Less affected by outliers/extreme scores

Standard Deviation
- Estimate of the average deviation of the scores from the mean
- Appropriate for interval and ratio level data
- Most robust measure of dispersion
- Corrected: used to estimate population standard deviation
- Uncorrected: used when you are not using the standard deviation to make estimates of the underlying population

24
Q

Week 6
Able to explain the terms: skew, kurtosis and bimodal distributions, and recognise how they apply to distributions of data

A

Skew
- The extent to which a frequency histogram is lopsided rather than symmetrical
- Suggests data that deviates from a symmetrical, ‘normal’ distribution

Kurtosis
- A measure of peak and flatness, or steep and shallowness
- Leptokurtic: higher kurtosis/very peaked
- Platykurtic: lower kurtosis/flat distribution
- Mesokurtic: between the two extremes

Bimodal Distributions
- 2 peaks in a distribution
- Suggests more than one distinct population

25
Q

Week 6
Familiar with the key features of a normal distribution

A
  • Peak in the middle
  • Tails off symmetrically at either side of the peak
  • Bell-shaped curve
26
Q

Week 6
Able to explain the standard normal distribution and its relationships to standard deviations and probabilities

A
  • Normal shaped distribution with a mean of 0 and a standard deviation of 1
  • A probability distribution - each score has a probability associated with it
  • The total area under the curve is 1 (/100%)
  • 2.5%, 13.5%, 34%, 34%, 13.5%, 2.5%
27
Q

Week 6
Aware of what z scores are, how to calculate them, their uses in research, and how they relate to standard normal distribution

A
  • Z score - tells us how many standard deviations above or below the mean a score is
  • Subtract the sample mean from the score, and then divide by the sample standard deviation
  • Uses: IQ, weight, relation to population mean
  • Allows us to compare scores from different samples, and different scores from the same samples
28
Q

Week 6
Appreciate the impact of outliers on data and recognise how outliers can be identified, e.g., box plots

A

Outliers
- Data points/scores that are very different to the rest of the data
- The mean is sensitive to extreme scores
- Outliers bias standard deviation and error estimates

Box Plots
- SPSS highlights any score it deems as outliers/extreme with a circle or star
- The box length is the IQR
- Outliers: circle, 1.5-3x the IQR
- Extreme: star, >3x the IQR

29
Q

Week 7
Describe confidence intervals and recognise their usefulness in estimating the population mean

A
  • We can use the sample mean as an estimate of the population mean
  • Confidence intervals: the boundaries within which we think the population value will fall
  • 95% confidence interval is typically used, especially in psychology - 1.96 SD above and below mean
  • 99.7% confidence is usually considered near-certain
30
Q

Week 7
Calculate standard error and confidence intervals

A

Standard Error
- Standard error = SD / square root of sample size

Confidence Intervals
- 1.96 standard deviations above and below the mean includes 95% of the standard normal distribution
- Calculate the standard error and the multiply by 1.96 for 95%

31
Q

Week 7
Recognise the purpose of hypotheses and their relevance to the psychological research process

A
  • Precise statement of an assumed relationship between variables
  • Must be testable
  • Research hypothesis is translated into a statistical null hypothesis (H0) for the purpose of statistical testing
32
Q

Week 7
Identify the different types of research hypothesis: causal, non-causal, directional, non-directional, and be able to give examples of each type of hypothesis

A

Causal
- Causal influence
- Only appropriate for experimental design
- E.g., consuming caffeine causes driving impairment

Non-Causal
- Suggests particular characteristics of behaviour without reference to causation
- E.g., consuming caffeine is associated with driving impairment

Directional
- Suggests the direction of the effect
- One-tailed hypothesis
- E.g., consuming higher levels of caffeine causes/is associated with greater driving impairment

Non-Directional
- Doesn’t specify the direction of the difference/effect
- Two-tailed hypothesis
- E.g., there will be a difference in driving impairment after consuming different strengths of caffeine

33
Q

Week 7
Appreciate the importance of the null hypothesis in statistical testing

A
  • Statistical tests are used to compare the null hypothesis to an alternative hypothesis
  • Can be disproven - theory is falsifiable
34
Q

Week 7
Describe what the p value is and how it relates to the null hypothesis

A
  • P value = probability value, how likely it would be to get the pattern of data we have found (or more extreme) if the null hypothesis were true
  • 5% (p: 0.05 or 1 in 20) is a cut-off point
  1. Formulate a hypothesis and collect data to measure this
  2. Run statistical analyses on the data (using SPSS, for example) to produce a test statistic
  3. Test statistic is then compared (in SPSS) with a known distribution of values of that statistics, that allows us to work out how likely it is to obtain if there were no effect in the population (if the null hypothesis were true)
  4. If p is less than 0.05, it suggests that the pattern of findings is unlikely to have arisen by chance - statistically significant
35
Q

Week 8
Be familiar with the term ‘inferential statistics’ and how it differs from ‘descriptive statistics’

A

Inferential
- Tells us something about the population based on the sample inferential statistics
- Aims at drawing conclusions

Descriptive
- Describe our sample(s)
- Uses means, median, mode, variance, frequencies

36
Q

Week 8
Be aware of the differences between parametric and non-parametric tests

A

Parametric
- Based on population parameters
- More assumptions
- Less universal
- Larger power (can detect effects when smaller samples are tested)

Non-Parametric
- Do not make any assumptions about the data distribution
- Fewer assumptions
- More universal - can always be applied
- Lower power (larger samples required to detect an effect)

37
Q

Week 8
Be able to identify when it is appropriate to conduct a t-test based on data

A
  • When we want to compare differences in means
  • Two separate groups
  • One group measured on two occasions
  • Whether one group differs from a specific mean
  • Scale the DV is measured on should be interval or ratio level
  • Data must be normally distributed
  • There should be no outliers/extreme scores
38
Q

Week 8
Recognise the differences between a repeated and an independent samples t-test

A

Repeated Measures
- Often used for before-and-after studies
- +: sample people - natural differences are controlled for
- Higher design power than independent samples
- SPSS:
– Paired samples statistics
– Paired samples correlation
– Paired samples test
– Paired samples effect size

Independent Samples
- Less sensitive than the repeated t-test
- SPSS:
– Group statistics
– Independent samples t-test
– Independent samples effect size

39
Q

Week 8
Be aware of the homogeneity of variance assumption

A
  • Must be checked for independent samples t-test
  • The variances of the populations should be approximately equal if comparing more than one group
40
Q

Week 8
Be familiar with how to interpret and report the results of a repeated and independent samples t-test (including effect sizes) from SPSS outputs

A

Repeated Measures
1. State what type of test has been performed and on what
2. Report the test statistic, df, statistical significance
3. Report the mean difference and associated confidence intervals
4. Report the effect size (Cohen’s d)
5. Comment on the means

Independent Samples
1. State what type of test has been performed and on what
2. Report the test statistic, df, statistical significance
3. Report the mean difference and associated confidence intervals
4. Report the effect size (Cohen’s d)
5. Comment on the means

41
Q

Week 9
Be familiar with what a correlation is used for and when it is appropriate to conduct a correlation on data

A
  • Finds a relationship between variables
  • Can suggest a what will happen to one variable as the result of a change in another
  • Can investigate the strength of a relationship
  • Positive/negative relationship

When to Conduct
- Data must be continuous (interval or ratio level)
- Data must be normally distributed
- No outliers/extreme scores
- Related pairs of data (every individual has scores on x and y)
- Linear relationship between variables

42
Q

Week 9
Be aware of the data assumption of linearity

A
  • The variables should be lineally related
  • Straight-line relationship between x and y
  • No relationship = no motive to run the correlation
43
Q

Week 9
Be able to interpret scatterplots and recognise their purpose in correlation

A
  • (Generally), IV on x axis; DV on y axis
  • Positive /, Negative \, Curvilinear n
44
Q

Week 9
Be familiar with the terms ‘direction’ and ‘strength’ in relation to correlation

A
  • Direction - positive/negative
    – -1 to +1
  • Strength - gradient of line
45
Q

Week 9
Be familiar with how to interpret and report the results of correlation and partial correlation (including effect sizes) from SPSS outputs

A
  • Square the correlation between variables to find out how much variance they share as a percentage
  • This would be the centre of a venn diagram - shared variance
  • Remaining percentage is unique variance
  • If the shared variance is greater than the unique variance, the correlation coefficient, r, will be high
  • Effect size - tells us how large the effect, or relationship, is
  • Small effect: r = 0.1
  • Medium effect: r = 0.3
  • Large effect: r = 0.5
46
Q

Week 9
Be familiar with how to interpret and report the results of correlation and partial correlation (including effect sizes) from SPSS outputs

A
  1. State what type of correlation has been performed, the variables correlated, and state the direction of the relationship found
  2. Report the test statistic, df, statistical significance (CI’s not generally reported for correlation)
  3. Report the effect size (Cohen’s r)
  4. Comment on the direction of the relationship
47
Q

(Week 9
Pearson Correlation: SPSS Outputs)
For the Pearson Correlation there will be:

A
  1. Descriptive statistics
  2. Correlations
  3. Confidence intervals
48
Q

Week 10
Recognise when it is appropriate to use parametric and non-parametric tests

A

Parametric
- Underlying probability distributions, e.g., normal distribution
- DV measured at interval or ratio level
- No outliers
- Homogeneity of variances (specific to independent t-test)
- Linearity (specific to correlation)

Non-Parametric
- ‘Distribution free’
- Ranks of data
- Outliers have little impact

49
Q

Week 10
Identify the alternative tests available for correlation and t-test that do not require parametric assumptions to be met: Spearman’s Rho, Mann Whitney U, Wilcoxon Signed-Rank

A
  • Non-normal distribution
  • Interval, ratio or ordinal data
  • Small sample size
  • Outliers
50
Q

Week 10
Give examples of the types of research question and hypotheses that would be appropriate for analyses using Spearman’s Rho, Mann Whitney U, and Wilcoxon Signed-Rank

A

Spearman’s
- Is there a relationship between student and teacher rankings of competence in maths?

Mann Whitney U
- Is there a difference between students’ enjoyment of a course based on the teaching and learning style? Do students prefer lectures or problem-based workshops?

Wilcoxon
- Is there a difference in body image ratings before and after viewing celebrity photographs on Instagram?

51
Q

Week 10
Be able to appropriately report in text the results of Spearman’s Rho, Mann Whitney U and Wilcoxon Signed-Rank from SPSS outputs

A

Spearman’s
- Outputs correlation
1. State what type of correlation has been performed, the variables correlated, and state the direction of the relationship found
2. Report the test statistic, df, statistical significance
3. Report the effect size (Cohen’s rs)
4. Comment on the direction of the relationship

Mann Whitney U & Wilcoxon
- Outputs ranks & test statistics
1. State what type of test has been performed and on what variables
2. Report the test statistic, z, statistical significance
3. Report the effect size (Cohen’s r)
4. Comment on the medians

52
Q

Week 11
Describe what internal and external validity are

A

Validity
- The extent to which an effect demonstrated in research is genuine

Internal - Avoid confounding variables

External - The degree to which results generalise beyond the experimental context, extent to which inferences can be made from sample to wider population

53
Q

Week 11
Recognise factors that threaten validity

A
  • Attrition/mortality
  • History
  • Sampling
  • Maturation
  • Testing and instrument issues
54
Q

Week 11
Describe what internal and external reliability are

A

Reliability
- The extent to which a measurements is reproducible or consistent over time

Internal - Internal consistency of the test

External - Test-retest, Inter-rater

55
Q

Week 11
Recognise the importance of reliability and validity in quanitative research

A
  • Credibility of research and scientific rigor
  • Ensure findings don’t conflict with each other
  • Clinical importance
56
Q

Week 11
Identify potential issues of reliability and validity posed by quantitative methods and designs

A

Experiments and RCTs
- Standardised procedures

Quasi Experiments
- Full control over the IV

Correlational Studies
- Non-manipulated variables

Questionnaires
- May not be valid - interpretation of questions