PUBH Final BLOCK ONE Actual questions and answers with 100% accuracy(VERIFIED BY PROFESSOR)

Chi-square (χ²)

Tests whether TWO categorical variables are associated/related. Use when BOTH variables are labels/groups/categories. Example: gender vs smoking yes/no. Trap: if one variable is a score/number, NOT chi-square.

Chi-square trap

Only use chi-square if BOTH variables are categorical. Example: gender + smoking yes/no = chi-square. Gender + anxiety score = NOT chi-square.

Independent samples t-test

Compares the average score between EXACTLY 2 groups. Use when one variable is categorical with 2 groups and the other is a continuous score. Example: men vs women anxiety scores. Trap: if 3+ groups, use ANOVA.

t-test trap

Only use t-test if there are EXACTLY 2 groups and one continuous score. Example: treatment vs control stress scores.

ANOVA (F-test)

Compares the average score between 3 OR MORE groups. Use when one variable is categorical with 3+ groups and the other is a continuous score. Example: freshman vs sophomore vs junior vs senior stress scores. Trap: if only 2 groups, use t-test.

ANOVA trap

Only use ANOVA if there are 3+ groups and one continuous score. Example: race groups vs anxiety score.

Pearson correlation (r)

Tests relationship between TWO continuous variables. Use when BOTH variables are numbers/scores. Example: stress score vs anxiety score. Trap: categories are NOT correlation.

Correlation coefficient (r)

Shows strength and direction of relationship between two continuous variables. Range = -1 to +1. Positive = same direction. Negative = opposite direction. Closer to 1 = stronger.

Positive correlation

As one variable increases, the other increases. Example: stress goes up, anxiety goes up.

Negative correlation

As one variable increases, the other decreases. Example: exercise goes up, depression goes down.

Correlation trap

Positive or negative does NOT change the test. It is still correlation if BOTH variables are continuous.

p-value (Sig.)

Used to determine statistical significance. In SPSS, Sig. = p-value. If p < .05 = statistically significant. If p > .05 = NOT statistically significant.

SPSS Sig. interpretation

Sig. = p-value. If Sig. < .05 = YES difference/relationship = reject null hypothesis. If Sig. > .05 = NO difference/relationship = fail to reject null hypothesis.

Memory trick for p-value

Small Sig = Big Deal. Small p-value means something statistically significant is happening.

Statistically significant

Result unlikely due to chance. Usually means p < .05.

Null hypothesis (H₀)

States NO difference, NO relationship, or NO association exists. Example: men and women have equal anxiety scores.

Alternative hypothesis (H₁)

States a difference, relationship, or association DOES exist. Example: men and women differ in anxiety scores.

Type I error

False positive. Rejecting the null hypothesis when the null is actually true. Example: saying treatment works when it does not.

Type II error

False negative. Failing to reject the null hypothesis when the null is actually false. Example: saying treatment does not work when it actually does.

Power

Ability of a study to detect a true effect if one exists. Higher power = better chance of finding significance. Common benchmark = .80.

Descriptive statistics

Used to summarize data. Examples: frequency, percent, mean, median, SD.

Inferential statistics

Used to test hypotheses between variables. Examples: chi-square, t-test, ANOVA, correlation.

Frequency (n)

Number of participants with a characteristic. Example: 30 smokers.

Percent (%)

Proportion out of 100. Example: 30% smokers.

Mean

Average score. Use with normally distributed continuous variables. Report with standard deviation. Sensitive to extreme values.

Median

Middle score. Use with skewed data or ordinal variables.

Mode

Most frequently occurring value.

Standard deviation (SD)

Shows spread of scores around the mean. Small SD = clustered scores. Large SD = spread out scores.

Quartiles

Q1 = 25th percentile. Q2 = median. Q3 = 75th percentile.

Interquartile range (IQR)

Q3 minus Q1. Used with skewed data.

Normal distribution

Bell-shaped symmetrical distribution. Report mean + SD.

Positive skew

Right tail. Most scores are low, few scores are very high. Example: income.

Negative skew

Left tail. Most scores are high, few scores are low.

Nominal

Categories with NO order. Examples: gender, race, smoker yes/no.

Ordinal

Ordered categories. Examples: Likert scale, pain levels, class rank.

Interval

Equal spacing but NO true zero. Example: temperature.

Ratio

Equal spacing WITH true zero. Examples: age, height, weight.

NOIR

Levels of measurement mnemonic: Nominal, Ordinal, Interval, Ratio.

Categorical variable

A variable made of labels/groups/categories. Examples: gender, race, yes/no.

Continuous variable

A numeric variable or score. Examples: age, anxiety score, stress score.

Scoring

Process of combining multiple questions into one score.

Composite score

A single score created from multiple items. Example: stress scale total score.

Reverse coding / reverse scoring

Flipping response values so all items point in the same direction. Example: 0→4, 1→3, 2→2, 3→1, 4→0.

Likert scale

Ordered response scale. Example: strongly disagree to strongly agree. Individual items are ordinal; summed scales may be treated as continuous.

Test statistic

The calculated number for a statistical test. Correlation = r, t-test = t, ANOVA = F, chi-square = χ².

How to choose the correct test

Categorical + categorical = chi-square. Continuous + continuous = correlation. 2 groups + continuous score = t-test. 3+ groups + continuous score = ANOVA.