NCE Research and Evaluation 2 — Questions and Answers

Question 1: A researcher finds a correlation of r = -0.75 between hours of sleep and self-reported anxiety scores. This finding means:

More sleep causes less anxiety
There is a strong negative relationship between sleep and anxiety (Correct answer)
Sleep and anxiety are unrelated
The relationship is moderate and positive

Correct answer: There is a strong negative relationship between sleep and anxiety

A correlation of -0.75 indicates a strong negative (inverse) relationship: as sleep increases, anxiety tends to decrease. However, correlation does not establish causation.

Pearson r values range from -1.00 to +1.00. Strength guidelines (Cohen): small = +/-0.10, medium = +/-0.30, large = +/-0.50. An r of -0.75 is very strong. The negative sign means as one variable increases, the other decreases. Critical caveat: correlational designs cannot determine causality -- other explanations include reverse causation (anxiety disrupts sleep) or a third variable (e.g., depression affects both).

Question 2: Which type of research design is BEST suited to establishing cause-and-effect relationships?

Correlational survey research
Case study methodology
True experimental design with random assignment (Correct answer)
Cross-sectional observational design

Correct answer: True experimental design with random assignment

True experimental designs with random assignment to conditions allow researchers to control extraneous variables and make causal inferences, because random assignment equalizes groups on known and unknown confounds.

Requirements for causal inference: (1) Temporal precedence (cause precedes effect), (2) Covariation (cause and effect vary together), (3) Elimination of alternatives (confounds ruled out). Only true experiments with random assignment satisfy all three criteria. Quasi-experimental designs lack random assignment and thus carry more threats to internal validity (selection bias, history effects, maturation). Correlational studies can only establish association, not causation.

Question 3: In counseling outcome research, a 'control group' serves which primary purpose?

To provide data that can be generalized to the population
To serve as a comparison point against which treatment effects can be evaluated (Correct answer)
To receive the most intensive form of the treatment being studied
To provide demographic data about the research sample

Correct answer: To serve as a comparison point against which treatment effects can be evaluated

The control group does not receive the experimental treatment and provides a baseline comparison. Differences between the treatment and control groups, after random assignment, can be attributed to the treatment.

Control group types in counseling research: No-treatment control, Waitlist control (participants receive treatment later, controlling for time effects), Active control/comparison (participants receive an alternative treatment, controlling for nonspecific factors like attention and expectation). The gold standard is the randomized controlled trial (RCT). Ethical concerns arise when withholding potentially effective treatment from distressed populations.

Question 4: Which measure of central tendency is MOST resistant to the influence of extreme outlier scores?

Mean
Mode
Median (Correct answer)
Standard deviation

Correct answer: Median

The median (the middle value when data are ordered) is resistant to outliers because it depends only on rank order, not the magnitude of extreme values. The mean is highly sensitive to outliers.

Example: Income data for a group of 5 people: $20K, $25K, $30K, $35K, $1M. Mean = approximately $222K (distorted by the millionaire); Median = $30K (better represents the typical person). The median is preferred for skewed distributions. Standard deviation is a measure of spread, not central tendency.

Question 5: A researcher uses a p-value threshold of 0.05 to determine statistical significance. This means:

There is a 5% chance the hypothesis is correct
There is a 5% probability of obtaining results this extreme if the null hypothesis is true (Correct answer)
The effect size is at least 5% larger than chance
The research findings apply to 95% of the population

Correct answer: There is a 5% probability of obtaining results this extreme if the null hypothesis is true

A p-value of 0.05 means there is a 5% probability of observing results at least as extreme as those obtained if the null hypothesis is actually true (i.e., a 5% Type I error rate).

Common misinterpretations of p-values: (1) p = 0.05 does NOT mean there is a 5% chance the results are due to chance; (2) statistical significance does not equal practical significance; (3) p > 0.05 does not prove the null hypothesis. Effect sizes (Cohen's d, eta-squared, r) must accompany p-values to convey the practical magnitude of findings. The APA Publication Manual recommends always reporting effect sizes alongside significance tests.

Question 6: A counselor educator conducts a study where participants complete questionnaires before and after a 10-week stress management program. This is an example of:

A cross-sectional design
A pre-test/post-test design (Correct answer)
A randomized controlled trial
A retrospective case study

Correct answer: A pre-test/post-test design

A pre-test/post-test design measures participants at baseline and again after an intervention to assess change. Without a control group and random assignment, it is a quasi-experimental design.

Pre-test/post-test designs are common in counseling outcome research and program evaluation. Internal validity threats without a control group: history (other events during the program caused change), maturation (participants improved naturally over time), testing effects (pre-test sensitization), regression to the mean (extreme scorers move toward the average regardless of treatment). Adding a waitlist control group and random assignment converts this to an RCT.

NCE Research and Evaluation 2

A researcher finds a correlation of r = -0.75 between hours of sleep and self-reported anxiety scores.
This finding means: