Research method modules

Introduction to psychological research methods

• Its always important to think critically about claims

• Psychology is an evidence-based discipline

• This evidence comes from research

• Research studies follow a process

Constructs

Intangible, abstract attributes that are theorised to underlie observable behaviour. They are not directly observable or measureable

Operational definition

The process of defining and measuring an unobservable construct indirectly

Research questions

The research process begins with a question

Research questions are broad ideas tha typically ask about either association, difference, or causation

Hypotheses

Logical, specific, testable, refutable, and predictive statements about what will happen in a psychological research study

Variables and measurement

• Variables are operationalised by using a measure in psychological research

• There are two types of variables: continous and discrete

• The scale of measurement is determined by the measure used

• There are four scales of measurement: nominal, ordinal, interval, and ratio

Continous variables

Allow for decimals/fractional values to be obtained between points on a scale. Eg distance, weight, time, etc

Discrete variables

Seperate indivisible categories, where values cannot meaningfully exist between points on a scale (whole numbers). Eg. number of people in a family

Nominal scale

• Involves measurement based on a set of categories (categorical).

• There is no ordering, and different values do not indicate differences in magnitude, they just indicate membership to different categories.

Ordinal scale

• Values are assigned to indicate an order

• The scale does not, however, tell you the magnitude of the difference between points on a measurement scale

• Eg. first, second, third place indicating order of arrival at the finish line in a running race. However, do not tell you the actual times/the difference between these times.

Interval scale

• Numbers indicate an actual amount of something present,

• Equal units of measurement seperate two scores on a scale

• There is not a real zero (0 doesn’t mean the absence of something - it’s arbitrary)

• Eg temperature indicates the amount of heat precent but 0 degrees celcius does not indicate that there is 0 heat.

Ratio scale

• Numbers indicate an actual amount of something

• Equal units of measurement seperate two scores on a scale

• There is a real 0

• Eg. if a ruler measures 0m, it indicates an absence of length/distance etc

Descriptive research design

• Concerned with the measurement and description of the natural state of individual variables as they are experienced by a certain group of people

• There is no experimentation and nothing is manipulated

• The focus is on observation only

• Eg measuring the heights of students in your class and making descriptions on what was typical by an average

• Eg measuring the prevalence of students sick with colds and/or flues in your class during the winter

Correlational research design

• Concerned with the investigation of the relationship between variables

• no experimental manipulation

• Variables are observed as they naturally exist

• and the presence of an association between the variables is assessed.

• Each participant in correlational research must provide information regarding two variables

• Relationship → the values on one variable are systematically and predictably accompanied by changes in another variable

• Cannot make determinations about cause and effect/ which variable influences the other in the relationship

• Could be a third variable that explains causation in the relationship.

• Eg. correlation between time spent exercising and happiness questionnaire score

Experimental research design

• Concerned with determining cause and effect in a relationship between variables

• Involves manipulation of the hypothesised independent variable in a relationship

• Involve carefully controlled experimental conditions to increase internal validity (degree of confidence in the relationship). Decreases the likelihood of confounding variables influencing results.

• The most common sampling is random

Quasi-experimental designs

• Investigate cause and effect relationships

• Usually the independent variable is a characteristic/demographic

• No random allocation to groups

• Limits ability to control for confounding variables

Non-experimental research designs

• Demonstrate the relationship between variables

• Do not attempt to explain cause and effect

• Involves the observation of two or more groups (eg arts and science students) and one variable (eg IQ)

• Different to correlational design as there is one variable being observed

• Focus on comparing groups

Population

Everyone of relevance to a research question/study

Probability sampling

• Population characteristics are known

• Eg quota sampling

Non-probability sampling

• Unknown population characteristics

• Eg convenience

NHMRC ethical considerations

• Merit

• Integrity

• Justice

• Beneficence

• Respect

Merit

• Research that is justified by its potential benefits to humanity

• Scientifically sound (likely to achieve its aims)

• Psychological researchers review relevant literature to asses the likely merit of a given study

Integrity

• The research has been undergone according to well founded principles of research conduct

• Research has been reported honestly and ethically

Justice

• Ensure that the inclusion and exclusion of research participants is fair and equitable

• No burden is placed on a particular participant group

• Participants are not exploited in any way

• There is fair and equitable access to the benefits of the research

Beneficence

• Risks and benefits are considerd

• Research should involve benefits to participants/community that should justify any risk or discomfort to participants

• When designing a study, it is imperative that any risks involved with participation are minimised and clearly explained to participants prior to their involvement with the study (informed consent)

Respect

• Respectful research practice holds respect for the value of research participants and of their culture, beliefs, and welfare.

• Researchers must respect the privacy and confidentiality of participants

• Must empower participants to feel control over their participation and the data they have contributed to the study (can remove themselves + data).

We can represent a frequency distribution in various ways, including:

• Frequency histogram

• Frequency table

• Boxplot

Central tendency

What is most representative in a distrubution

Variability

How scores in a distribution differ or not

Measures of variability

• Range

• IQR

• Standard deviation

Standard deviation for a sample

The average amount that scores differ or deviate from the mean

Calculating standard deviation

1. Sum of squares (solves problem of negative scores)

2. Variance (sum of squares divided by one less than sample size)

3. Standard deviation (square root of the variance)

Deductive argument

• Valid

• Conclusion necessarily follows from premise/s

• Broad (eg theory) → specific (eg hypothesis)

Inductive arguments

• Strong

• Premises provide good support for the conclusion

• Specific observation (eg study results) → general explanation (eg theory)