Psychology: Data analysis & research method test

Quantitative data

Numerical data that is obtained through standardized measurements or counting. Used for data analytics, comparison, and identifying patterns.

Qualitative data

Descriptive, non-numerical data that is used to capture experience, opinions and observable behaviours.

Operationalisation of data

The explicit, precise definition of how an abstract psychological concept is manipulated or measured in a specific study.

Independent variable (IV)

The variable that is being manipulated by the researcher to see an effect on the dependent variable.

Dependent variable

The outcome factor that the researcher measures in response to the manipulation of the IV.

Confounding variables

Variables that vary alongside with the independent variables that unintentionally influence the dependent variables, leading a false impression on the cause-and-effect relationship.

Primary data

First-hand empirical data that is collected directly by the researchers to address their unique question.

Secondary data

Pre-existing data that are collected by previous researchers, organizations or institutions for secondary research purposes.

Empirical data

Scientific evidence acquired by systematic observation, measurement or controlled experimentation.

Anedotal evidence

Data that is based on personal experience, stories or isolated instances.

Self reported data

Subjective data provided directly by participants regarding their opinions, behaviours or thoughts.

Behaviour data

Objective measurements of directly observable of participant’s action or task performance.

Biological data

Physiology data that is measured via biological instrument.

Archival data

Pre-existing public or organizational records that is analyzed for research purpose.

Cross sectional study

An observational design that compares different groups of participants at one specific time.

Longtitudinal study

An observational design that tracks and measures the same group of participants over time.

Predict, Observe, Infer (POI)

Predict: stimulate theoretical expectations, Observe: collect and observe raw data without interpretation, Inference: draws psychological conclusions and evaluate limitations.

Nominal scale

Data are discrete or named categories that have no numerical value — mostly analyzed by mode - the most frequent data

Ordinal scale

Data are placed into ranked or ordered where relative position is known — mostly analyzed by IQR, Median

Ratio scale

Numerical data with equal intervals and an absolute value equal 0, a value = 0 indicates the absence of what is being measured — mostly analysed by mean, standard deviation.

Central tendency

A statistical value that represents the typical or central value of a dataset (mean, median, mode)

Mean (Artithemic average)

The sum of all numerical values is divided by the total number of scores (N)

Median (middle score)

The middle score in an ordered dataset, mostly used for ordinal data or skewed distributions.

Mode

The single score or the category that appears with the highest frequency.

Disperation

A statistical metric that quantifies the spread, scatter, or variability of a dataset.

Range

The difference between the highest score and the lowest score - can be distorted by a single extreme outliers

Standard deviation

Calculate the average distance for every individual score from the mean of its dataset - high SD —> high spreading out.

Interquartile range (IQR)

The spread of the middle 50% of the scores, measured by Q3-Q1.

Normal distribution (Bell shaped curve)

A symmetrical, bell-shaped theoretical frequency curve where mean, median and mode are the same at the middle point.

Positive skewness/colleration

A distribution asymmetrical toward the low end, leaning torward the postive values.

Negative skewness/colleration

A distribution asymmetrical to the high end, leaning towards the negative values of the dataset.

Bar graph

A visual representation that places discrete, categorized data (IV) alongside continuous data (DV)—the bars must be separated and untouched.

Histogram

A graph showing the frequency distribution of continuous numerical grouped data into intervals (bin) with bars touching each other.

Modal class

A specific interval (bins) on the histogram that contains the highest frequency of occurences.

Box&Whisker plot

A visual chart displaying the summary of 5 values: Min, Q1, Median (Q2), Q3, and Max - it also displays IQR and extreme outliers.

Outlier

An extreme data point that differs significantly from the values on the dataset.

Scatter plot

A graph showing individual data points on the X and Y axes continuously to evaluate the strength of their correlation.

Truncated Y-axis rule

The Y-axis must always start from 0 to avoid exaggerating of the difference between groups.

Null hypothesis (H0)

A testable scientific statement states that there’s no relationship between the IV and the DV.

Alternative/Research hypothesis

Predicts that there’s a statistical difference, effect, and relationship between variables.

p - Value (probability value)

The probability of obtaining results as extreme as the observed results, assuming that the null hypothesis is true.

Standard significance threshold (p ≤ 0.05)

Results are significant if p ≤ 0.05 —> rejects H0 and accept H1

High significance threshold (p < 0.01)

Less than 1% of obtaining the observed results by random chance if H0 is true —> higher chance to reject H0

Type I error (False positive)

Incorrectly rejecting the null hypothesis - scientists claim that there’s an effect exists when it is due to luck.

Type II error (False negative)

Incorrectly falling the null hypothsesis - scientists claim that there’s no effect when there’s actually is.

Correlation

When two variables are associated or change together, but one does not need to affect another.

Causation

Cause and effect relationship - when one variable cause the change in the other variable.

Bidirectional Ambiguity

A limitation in correlation study that makes it impossible to identify which variable causes the change in another, because either one of them can.

Third variable problem

When an unmeasured factor influences two other variables, creating a correlation that does not nessecary invole the cause and effect relationship.