Research Methods and Statistics

Scientific Evidence

Evidence which serves to either support or counter a scientific theory or hypothesis.

Theory

An attempt to understand and explain something.

Induction

Associated with qualitative research.

Deduction

Associated with quantitative research.

How can induction and deduction work together?

Through induction, researchers observe facts and formulate theories. Then, through deduction, researchers can test the theories.

Explanatory Power

The ability to explain a phenomenon effectively.

Falsifiability

A principle that determines whether a claim can be proven false, central to scientific inquiry as argued by Popper.

According to Mayo (2018) when has a claim been ‘severely tested’?

A claim is severely tested to the extent it has been subjected to and passed a test that probably would have found flaws, were they present.

HARKing

The practice of only reporting results that are relative to the hypothesis/study while ignoring non-significant results.

What is the significance of hypothesis testing?

To test whether the null hypothesis can be rejected or approved. If the null hypothesis is rejected, then the research hypothesis can be accepted.

Measures of Centre

Values that describe the middle or central characteristics of a set of data.

Mode

The value that appears most frequently in a data set.

Median

The middle value when a data set is ordered from least to greatest.

Mean

The average of a set of numerical values, calculated by summing the values and dividing by the count.

Know the relationship between the measures of centre in symmetrical vs skewed distributions.

When the dat are distributed normally, the three measures of centre all fall at or near the same value. But when the data are skewed, these measures are no longer identical.

Independent Variable (IV)

A variable of interest that potentially causes changes in the dependent variable.

Dependent Variable (DV)

The variable used to measure or assess the outcome on each individual in a study.

Extraneous Variable

Any variable that is potentially associated with the response variable but not of main interest.

Nuisance Variable

An unwanted variable that correlates with the dependent variable but is not of interest to the researcher.

Confounding Variable

A variable that obscures the observed relationship between the response and explanatory variables.

Quantile

A statistical measure that divides a dataset into equal-sized, ordered groups.

Common quantiles

Tertiles, quartiles, quintiles, deciles, percentiles

Skewness

A measure of the bilateral symmetry of the data.

Kurtosis

A measure of how heavy the tails of the distribution curve are and how sharp or flat the peak is.

Internal Validity

A measure of control within the study ensuring that the results are due to the independent variable.

External Validity

The ability to generalize the results of an experiment to the wider population or context.

Two types of external validity

Population, ecological

Sampling error

The amount of error in the estimate of a population parameter that is based on a sample statistic.

How the changes in the level of control effect both internal validity and external validity?

Higher control will result in higher internal validity, lower external validity.

Operational definition

Define exactly how something will be identified, measured, observed or assessed.

Proxy

An indirect measure of the desired outcome which is itself strongly correlated to that outcome.

PICO components

Population, Intervention, Control, Outcome used in formulating research questions.

Null Hypothesis (H0)

A hypothesis that states there is no effect or no difference; the default assumption to be tested.

Alternative Hypothesis (H1)

A hypothesis that represents the expected effect or difference in a study.

Statistical Significance

A determination that results are not likely due to chance, typically assessed using p-values.

P-value

A descriptive continuous measure indicating the compatibility between observed data and the null hypothesis.

Alpha Level

A pre-determined threshold below which results are considered statistically significant.

Blinding

The practice of keeping participants and/or researchers unaware of treatments to reduce bias.

Placebo

A non-effective treatment used in research to control for placebo effects.

Control Group

A group in an experiment that does not receive the treatment, allowing for comparison.

Observer Effect

The influence of a researcher's expectations on participant behavior, which can be minimized by blinding.

Carry-over Effect

When past treatments influence responses to current treatments, minimized by washout periods.

Washout Period

A period during which participants do not receive any treatment, used to prevent results from mixing.