Final Exam | Research Methods

A researcher sets alpha at 0.05 and obtains p = 0.03. Which of the following is the correct conclusion?

Reject H₀; there is a statistically significant difference

A study reports p = 0.04 (alpha = 0.05) and concludes that a new exercise reduces knee pain. In reality, the exercise has NO effect. This is an example of:

Type I error (false positive)

Statistical power is best defined as:

The probability of detecting a real effect if one truly exists

A physical therapy study recruits 12 participants with no power analysis. Which error type risk is MOST increased?

Type II error, because the study may miss a real effect

A researcher measures pain sensitivity and physical activity in 200 adults at a single clinic visit. Which study design does this represent?

Cross-sectional observational study

A researcher wants to determine if a new manual therapy technique CAUSES a reduction in chronic low back pain. Which study design best supports a causal claim?

Randomized controlled trial

Which of the following distinguishes an experimental study from a quasi-experimental study?

Experimental studies use random assignment; quasi-experimental studies do not

In an RCT on HIIT and pain, a researcher excludes participants currently receiving physical therapy. The PRIMARY reason for this exclusion is to:

Prevent confounding of the intervention effect

A dataset of patient pain scores is: 2, 3, 4, 5, 6, 7, 40. Which measure of central tendency is MOST appropriate to report?

Median, because it is resistant to the outlier (40)

A study reports: grip strength mean = 46 kg, SD = 10 kg (normal distribution). Approximately what percentage of participants had grip strength between 26 kg and 66 kg?

95%

Two clinics report the same mean pain reduction of 3 points. Clinic A has SD = 0.5; Clinic B has SD = 4.0. Which statement is MOST accurate?

Clinic A's outcomes are more consistent; Clinic B has much greater variability

When a researcher uses the Shapiro-Wilk test and obtains p = 0.08, the appropriate next step is to:

Assume normality and proceed with parametric tests

A researcher hypothesizes that HIIT reduces pain more than stretching. The correct null hypothesis (H₀) is:

There is no statistically significant difference in pain reduction between HIIT and stretching

A study on HIIT for chronic pain excludes participants with active cardiac conditions. This exclusion is primarily motivated by:

Participant safety — protecting those at risk of adverse events

A researcher recruits participants for a pain study by posting flyers in a PT clinic waiting room and taking whoever volunteers first. This is BEST described as:

Convenience (non-probability) sampling

A pain study is conducted at a sports medicine clinic and recruits 78 patients with knee OA. The TARGET POPULATION is best described as:

All adults with knee OA in the relevant population of interest

Which of the following is MOST likely to increase the risk of a Type I error (false positive)?

Setting alpha at 0.20 instead of the standard 0.05

According to the principles of a good research question, which criterion ensures that findings are relevant and impactful to a large number of patients?

Significance

A research paper's Table 1 shows: "Mean age = 45.2 (SD = 8.1), 62% female, mean pain score = 6.4 (SD = 1.9)." This table represents:

Descriptive statistics

A researcher uses hospital electronic health records to compare rates of opioid use in patients with vs. without prior physical therapy for low back pain. This is BEST classified as:

Retrospective observational study

A dataset of patient wait times has many short waits and a few extremely long waits. The distribution curve has a long tail extending to the RIGHT. This is best described as:

Positive (right) skew

In a positively skewed distribution, which of the following correctly describes the relationship between the mean and median?

Mean > Median

A study reports grip strength: mean = 50 kg, SD = 5 kg, normally distributed. Approximately 95% of participants had grip strength between:

40-60 kg

Using the same dataset (mean = 50 kg, SD = 5 kg), what is the score that is 2 standard deviations BELOW the mean?

40 kg

A researcher records pain scores for 7 patients: 2, 5, 3, 8, 6, 4, 7. What is the median pain score?

5

A dataset is: 1, 3, 4, 5, 6, 7, 7, 8, 8, 8, 9, 9, 50, 51. What is the median?

7.5

A clinician is evaluating five sources of information. From LOWEST to HIGHEST level of evidence, the correct ranking is:

Expert opinion → Case reports → Cohort → RCT → Meta-analysis

Before conducting a clinical study, a PT researcher submits her protocol to the Institutional Review Board (IRB). The PRIMARY purpose of this review is to:

Protect the rights and welfare of human research participants

A manuscript is submitted to a journal and undergoes peer review. The PRIMARY purpose of this process is to ensure that:

The study is methodologically and scientifically sound

Evidence-Based Practice (EBP) integrates three key components. Which of the following BEST describes the three pillars of EBP?

Best available research evidence + clinical expertise + patient values/preferences

A study table shows: Group A — Mean BMI = 25.96, SD = 1.00 (normal distribution). Approximately 95% of Group A participants had a BMI between:

23.96 - 27.96 (Mean +/_ 2 SD)

A study reports: Mean pain score = 6.4, SD = 1.9 (normal distribution). Which range contains approximately 95% of pain scores?

2.6 - 10.2 (Mean +/_ 2 SD)

Which of the following study types is typically RETROSPECTIVE (looks backward at already-collected data)?

Case-control study

Which statement BEST describes evidence-based practice (EMP)?

The conscientious, explicit, and judicious use of current best evidence in decisions about individual patient care

A strong research question should have ALL of the following characteristics EXCEPT:

It is a simple replication of a question that has already been answered

Which statement about the two major categories of statistics is CORRECT?

Descriptive statistics summarize/ describe the sample; inferential statistics draw conclusions about the population

A study analyzes data collected at a single point to provide a 'snapshot' of characteristics or associations. This design is:

Cross-sectional study

Which qualitative research design seeks to understand the 'lived experience' of individuals with a particular concept, situation, or event?

Phenomenological research

Which qualitative design seeks to understand a specific process or action by generating a theory, using methods such as interviews with open, axial, and selective coding?

Grounded theory

A researcher enrolls the first 30 patients who walk into the clinic and agree to participate. This sampling method is BEST described as:

Convenience (non-probability) sampling

A study is described as 'underpowered.' The most likely problem and its consequences are:

The sample is too small, increasing the risk of Type II error

The following statements about variables are all true EXCEPT:

Only the dependent variable can have levels

A confounder is a third variable that is related to BOTH the independent variable and the dependent variable and can distort the true relationship between them.

True

A study finds that patients who exercise more use fewer opioids. However, patients with lower baseline pain severity BOTH exercise more AND use fewer opioids. In this study, pain severity is best described as a:

Confounder

'Adjusted' results, such as an adjusted odds ratio or adjusted beta coefficient, indicate that the analysis controlled (accounted) for potential confounders.

True

Ten bilingual clinical instructors each independently rate the SAME 25 students' Spanish proficiency, and the researchers then compare the instructors' scores with one another. What type of reliability is being tested?

Inter-rater reliability

A single full-time research assistant will collect ALL of the pain-sensitivity data for a study using one device. Which type(s) of reliability must be ensured?

Intra-rater (and re-test of the instrument)

Face validity means that a measurement, on the surface, appears to assess what it is intended...

True

Construct validity evaluates whether a test truly represents the underlying concept (construct) that the researcher intends to measure.

True

The following statements about validity are all INCORRECT, EXCEPT:

Concurrent validity refers to the correlation of a new test with an established measure administered at the same time

A researcher estimates overall body-fat percentage by tape-measuring wrist circumference. Assistants are trained so that each gets the same result every time and their results agree with one another. The BEST description of this situation is:

The test is reliable but does not appear to have face validity

A researcher records gait speed (meters per second) for 200 older adults. What level of measurement is gait speed?

Ratio

Course satisfaction is rated as 'extremely satisfied,' 'satisfied,' 'not satisfied,' and 'will never take this course again!!' What is the level of measurement?

Ordinal

Participants are categorized as 'grumpy,' 'cocky,' or 'short-tempered.' What is the level of measurement for this variable?

Nominal

The following statements about levels of measurement are all true EXCEPT:

Temperature in degrees Celsius is ratio

A confidence interval (CI) provides information about precision and the range of plausible population values, whereas a p-value alone only tells you whether a result is statistically significant.

True

The following statement about p-values and 95% confidence intervals are all true EXCEPT:

Increasing the sample size makes the 95% CI wider

Which of the following statements about effect size is INCORRECT?

Effect size is the ability to detect a difference or relationship if one truly exists

You can reject the null hypothesis in all of the following scenerios EXCEPT:

The 95% CI for a beta coefficient includes 0

If the 95% CI of an odds ratio includes 0, then the p-value is greater than alpha.

False

It is recommended to use non-parametric statistical methods in all of the following situations EXCEPT:

The outcome is interval-level with a normal distribution

A researcher measures a 0-100 mental-health score BEFORE and AFTER treatment in the SAME 100 health-care providers to see if treatment improved the score. The most appropriate statistical test is:

Paired t-test

A researcher wants to compare the mean weekly exercise hours across THREE cohorts (Class of '25, '26, and '27). The most appropriate statistical test is:

ANOVA

You analyze a single DPT cohort's pre-test and post-test scores on an anatomy exam. This requires a statistical test that allows for the _____ comparison.

Within-group

A significant ANOVA result tells you that:

At least one group differs significantly, but NOT where the difference is (post-hoc testing is needed)

You measure running speed on a treadmill that maxes out at 5 mph, but several participants are elite college sprinters (mean male speed ≈ 8 mph). What measurement problem will you likely encounter?

Ceiling effect

The following statements are all true EXCEPT:

Statistical significance and clinical significance are the same thing

A correlation of r = -0.57 is found between anxiety scores (higher = more anxious) and physical-activity scores (higher = more active). The correct interpretation is:

Higher anxiety is associated with lower physical activity, AND lower anxiety with higher physical activity

Which statement about correlation analysis is CORRECT?

Correlation just describes a bivariate relationship

Examine the correlation matrix below. Which statement is TRUE?

Variable | 1 | 2 | 3

1. Sleep Quality | 1 |

2. Depression | -0.42** | 1

3. Physical Activity | 0.38** | -0.15 | 1

**Correlation is significant at p < 0.01

There is a significant negative correlation between sleep quality (#1) and Depression (#2)

In a linear regression, the beta for the effect of daily step count (per 1,000 steps) on a 0-10 pain score is -0.35 (95% CI: -0.60, -0.10). The correct interpretation is:

For every additional 1,000 steps, pain decreases by 0.35 points, and this is significant because the CI excludes 0

The table shows adjusted odds of long-term opioid use by post-op PT dose. Which interpretation is CORRECT? (OR=odds ratio)

Post-op PT dose | Adjusted OR (95% CI)

≥13 sessions | 0.80 (0.65, 0.99)

6-12 sessions | 0.92 (0.74, 1.15)

1-5 sessions | 1 [Reference]

Compared with the reference group (1-5 sessions), ≥13 sessions had ~20% reduced odds of long-term opioid use, and it is significant (CI excludes 1)

A logistic regression reports an OR of 0.54 (95% CI: 0.34, 0.88) for the effect of a one-unit increase in a pain-facilitation score on achieving a good surgical outcome. The correct interpretation is:

For every one-unit increase, there is a 46% reduced odds of a good outcome, and it is significant