1/23
Vocabulary flashcards covering descriptive and inferential statistics, normality testing, independent and paired-samples t-tests, and SPSS data screening based on PS1007 coursework materials.
Ratio Level of Measurement
A measurement scale that has equal intervals between values and an absolute, true zero point where zero represents the complete absence of the property (such as reaction time in milliseconds, where 0ms indicates no time elapsed).
Median
The measure of central tendency that best represents the typical score in a strongly skewed distribution (such as household income) because, unlike the mean, it is not distorted or pulled by extreme values.
Positively Skewed Distribution
A distribution where most scores cluster at the lower end of the scale and the tail stretches out toward the right (the positive end).
Inferential Statistics
Statistical procedures used to determine whether observed differences between sample conditions are larger than what would be expected by chance, going beyond descriptive statistics which only summarize sample properties.
Degrees of Freedom (df)
The number of independent values free to vary in a statistical calculation, reported in parentheses immediately following the test statistic (such as 58 in t(58)=−3.72).
p-value
The probability of obtaining test results at least as extreme as those observed in the sample, under the assumption that the null hypothesis (H0) is true.
Failing to Reject the Null Hypothesis
The statistical conclusion made when p>.05, meaning there is no evidence of a difference; this does not prove that no difference exists, and researchers never accept or prove the null hypothesis.
Type I Error
A false positive error occurring when a researcher rejects the null hypothesis and concludes there is an effect or difference when in reality the null hypothesis is true.
Two-Tailed Test Rejection Region
A significance testing region for an alpha level of α=.05 that divides the 5% total rejection area equally between both ends of the sampling distribution (2.5% in each tail).
Shapiro-Wilk Test
A statistical test used to evaluate normality, typically applied when a group has fewer than 50 observations; a significance value of p≤.05 indicates that the data deviate significantly from a normal distribution.
Standard Error of the Mean (SEM)
A measure of the variability of sample means that decreases as sample size increases, because larger samples allow unusual scores to balance out so that sample means vary less and sit closer to the true population mean.
Paired-Samples t-Test Normality Assumption
The assumption that the difference scores between conditions (condition A minus condition B) are approximately normally distributed, rather than the separate distributions of each condition.
Within-Subjects Design Power Advantage
The statistical advantage of repeated-measures designs where each participant acts as their own baseline, cancelling out individual differences and reducing unsystematic noise.
Independent-Samples t-Test
A parametric test used to compare the means of two separate, independent groups on a continuous dependent variable when the distribution within each group is approximately normal.
Degrees of Freedom (df) for Independent-Samples t-Test
The degrees of freedom calculated as total sample size minus two (df=total N−2), such that two groups of 25 participants yield df=50−2=48.
Levene's Test for Equality of Variances
An inferential test assessing whether two groups have equal variances; if p≤.05, the assumption of equal variances is violated, requiring the researcher to interpret the 'Equal variances not assumed' row of SPSS output.
Signal-to-Noise Ratio in the t-Statistic
The principle that t=noisedifference; for a given mean difference, a sample with smaller within-group variability (smaller standard deviation) produces less noise, resulting in a larger t-statistic and a smaller p-value.
Counterbalancing
An experimental technique used in repeated-measures studies where participants are split so that half receive condition order A-then-B and the other half receive B-then-A, preventing order or carryover effects.
Five-Step APA Results Write-Up Structure
The five required reporting components for an inferential test paragraph: 1) Justify (state normality checks), 2) Name (identify the test conducted), 3) Descriptives (report means and standard deviations), 4) Inferential (report test statistics: t, df, and p), and 5) Meaning (explain practical conclusions).

SPSS Tests of Normality Output
An SPSS output table reporting Kolmogorov-Smirnov and Shapiro-Wilk tests; a non-significant Shapiro-Wilk result (such as p=.999 with df=24) indicates that difference scores do not differ significantly from a normal distribution.

Histogram for Normality Assessment
A visual distribution assessment tool where a roughly bell-shaped, symmetrical appearance with a single central peak and no extreme scores provides evidence that difference scores follow a normal distribution.

Normal Q-Q Plot
A probability plot comparing observed values against expected normal values; data are consistent with normality when points lie closely along the straight diagonal line with minimal departure.

Boxplot for Normality Assessment
A graphical representation showing data symmetry and spread; consistency with normality is indicated when the median line sits near the center of the box, whiskers are approximately equal in length, and no outlier data points appear.
Statistical Significance
A finding where the probability value falls below the predetermined threshold (p<.05, such as p=.021 or p=.002), leading to the rejection of the null hypothesis in favor of the alternative hypothesis.