PSYC10003 · Mind, Brain and Behaviour 1
Research Methods and Statistics for Psychology
This chapter consolidates the twelve self-directed Research Methods modules — the fourth exam strand, worth 25% of the paper. It covers constructs and operational definitions, hypotheses, variables and the scales of measurement, study design, sampling, research ethics, distributions, descriptive statistics (central tendency and variability) and critical thinking, plus the JASP-based practical work. The examinable skills are classifying a measurement on its scale and variable type, choosing the right descriptive statistic and figure, and interpreting JASP output — the strand many students find the most learnable 25%.
What this chapter covers
- 01Constructs and operational definitions; research questions vs testable hypotheses
- 02Variable type (discrete vs continuous) and the four scales (nominal, ordinal, interval, ratio)
- 03Study design: descriptive, correlational, experimental (manipulate IV + random assignment), quasi-experimental
- 04Correlation does not imply causation; sampling (probability vs non-probability) and representativeness
- 05Research ethics: the NHMRC values (merit, integrity, justice, beneficence, respect); informed consent
- 06Distributions and appropriate figures: histogram/boxplot (continuous) vs bar chart/frequency table (categorical)
- 07Central tendency (mode/median/mean) and skew; variability (range, IQR, SS, variance, SD)
- 08Reading JASP output: variable-type icons and choosing statistics by scale
Classifying measurements and choosing the right statistic
- +1(a) Number of lectures attended is a count: ratio (a true zero means none; equal intervals) and discrete (no half-lectures). Continuous-data figures apply loosely, but as a count use a bar chart/frequency table or histogram; the mean is appropriate if roughly symmetric (median if skewed).
- +1(b) A 0-10 rating where 0 = 'very low' (not 'none') has equal-ish spacing but no true zero, so it is interval and treated as continuous → histogram/boxplot; report the mean (median if skewed).
- +1(c) Faculty is an unordered category set: nominal and discrete → bar chart / frequency table; the only usable central-tendency measure is the mode.
- +1(d) An ordered five-level band is ordinal and discrete (ordered but with unequal/undefined gaps) → bar chart / frequency table; report the median or mode, not the mean.
Key terms
- Operational definition
- A definition of an intangible construct in specific, measurable terms — how it will be observed or measured in a study, turning a construct into a scorable variable.
- Scales of measurement
- Four levels of increasing information: nominal (unordered categories), ordinal (ordered, unequal gaps), interval (equal gaps, no true zero) and ratio (equal gaps and a true zero).
- Experimental design
- A design that manipulates an independent variable and uses random assignment and control of extraneous variables to establish causation — distinguished from correlational designs, which only measure association.
- Central tendency
- What is most typical in a distribution: the mode (most frequent), the median (middle value, robust to skew and outliers) and the mean (arithmetic average, sensitive to skew and outliers).
- Standard deviation
- The approximate average distance of scores from the mean, equal to the square root of the variance and expressed in the original units; paired with the mean for symmetric data (the IQR pairs with the median).
- NHMRC ethical values
- The five values governing Australian human research — merit, integrity, justice, beneficence and respect — operationalised through informed consent, voluntary participation, confidentiality and debriefing.
Research Methods and Statistics for Psychology FAQ
How do I tell interval from ratio?
Ask whether zero means 'none'. A ratio scale has a true zero (zero errors, zero minutes) so ratios are meaningful; an interval scale has an arbitrary zero (a 0-10 rating where 0 means 'very low', or temperature in degrees Celsius) so a score of 0 is a point on the scale, not absence. This distinction is one of the most common Research Methods MCQs.
Which central-tendency and variability measures go with which data?
For roughly symmetric interval/ratio data, report the mean with the standard deviation. For skewed data or data with outliers, report the median with the interquartile range, because the median and IQR are robust to extremes. For nominal data only the mode is meaningful. Matching the statistic to the scale earns easy marks.
What makes a study a true experiment?
Two features: the researcher manipulates an independent variable, and participants are randomly assigned to conditions (with extraneous variables controlled). Merely measuring whether two variables are related is a correlational design, which cannot establish causation — correlation does not imply causation is a recurring exam point.
Why is Research Methods examinable if it is self-directed?
Because it is one of the four equally-weighted exam strands, worth 25% of the 100-item paper, and the module pages state it is assessed on the final exam. The practical classes (Pracs 7-8) train the JASP interpretation skills the exam draws on. Many students find it the most learnable 25% because the rules are concrete.
Will I have to interpret JASP output?
Yes — the Research Methods block can show a JASP descriptives table or figure and ask you to identify the variable type from its icon (ruler = interval/ratio, bar chart = ordinal, three circles = nominal), pick the appropriate statistic or figure, or read skew from the mean-median relationship. No arithmetic is required beyond simple reading, but you must know which statistic suits which scale.
Exam move
Treat Research Methods as the most bankable 25% because its rules are concrete. Master the measurement-classification decision (true zero → ratio vs arbitrary zero → interval; ordered → ordinal vs unordered → nominal) and immediately pair each scale with its correct figure and statistic. Learn the central-tendency and skew relationships (mean = median = mode symmetric; mean > median positive skew) and be able to run the SS → variance → SD chain and read a JASP icon. Memorise the five NHMRC ethical values and the definition of a true experiment. Do not skip the self-directed modules during SWOTVAC — they are examinable and quick to bank. Confirm exam details on Canvas.
Working through Research Methods and Statistics for Psychology in PSYC10003? Sia is AskSia’s AI Psychology tutor — ask any PSYC10003 Research Methods and Statistics for Psychology question and get a clear, step-by-step explanation grounded in how PSYC10003 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
