For the AIIMS Physiotherapy entrance exam (and other competitive exams like JAMIA, PGI), Research Methodology and Biostatistics typically carry 10-15% of the weightage. These questions are usually conceptual. Below is a simplified, exam-focused breakdown.
1. Scales of Measurement
Remember the mnemonic NOIR (Nominal to Ratio) to recall the hierarchy from lowest to highest precision.
| Scale | Keywords/Definition | Physiotherapy Examples |
|---|---|---|
| Nominal | Names, Categories, No Order. Only counting is possible. | Gender (Male/Female), Blood Group, Type of CP (Spastic/Ataxic). |
| Ordinal | Ordered categories, Ranks. Differences between ranks are unknown/unequal. | MMT Grades (0-5), Ashworth Scale, Glasgow Coma Scale, Pain Severity (Mild/Mod/Severe). |
| Interval | Equal intervals, but NO Absolute Zero (Zero is arbitrary). | Temperature (Celsius/Fahrenheit), IQ Scores. |
| Ratio | Highest level. Has a True Absolute Zero. Ratios are meaningful. | Range of Motion (Goniometry), Height, Weight, Walking Speed, Age. |
2. Measures of Central Tendency
Mean (Average)
- Most commonly used measure.
- Formula: Sum of observations / Number of observations.
- Drawback: Highly affected by extreme values (outliers).
Median (Middle Value)
- The middlemost value when data is arranged in order.
- Best Use: When data is Skewed or has outliers.
- Example: In income data or recovery time, Median is better than Mean.
Mode (Most Frequent)
- The value that repeats the most.
- Best Use: For Nominal data (e.g., "What is the most common type of shoulder injury?").
3. Sampling Methods
Probability Sampling (Random)
- Simple Random Sampling: Lottery method. Everyone has an equal chance.
- Stratified Random Sampling: Population is divided into subgroups (Strata) based on characteristics (e.g., Age, Gender), and then random samples are taken from each stratum. Best for representation.
- Cluster Sampling: Population is divided into clusters (usually geographic), and whole clusters are selected.
Non-Probability Sampling
- Convenience Sampling: Selecting patients who are easily available (e.g., patients visiting your specific OP clinic). Most common in clinical research but has high bias.
- Purposive Sampling: Researcher selects specific subjects based on criteria.
4. Tests of Significance (Kab use karte hain?)
This is the most confusing part for students. Use this cheat sheet:
| Test Name | When to use it? | Example |
|---|---|---|
| T-Test (Student's t-test) | Comparing Means of 2 Groups. | Comparing ROM between Ultrasound Group vs. SWD Group. |
| Paired T-Test | Comparing Means of the SAME Group (Before vs. After). | VAS score Pre-treatment vs. Post-treatment in the same patient. |
| ANOVA (Analysis of Variance) | Comparing Means of More than 2 Groups (>2). | Comparing pain relief among Group A (IFT), Group B (TENS), and Group C (Placebo). |
| Chi-Square Test | Comparing Categorical/Proportions data (Non-parametric). | Association between Gender (M/F) and Frozen Shoulder (Yes/No). |
| Pearson Correlation (r) | To check the relationship between two continuous variables. | Relationship between BMI and Knee Pain intensity. |
5. Research Designs
Randomized Control Trial (RCT)
- The Gold Standard for experimental studies.
- Must have: Randomization + Control Group + Manipulation (Treatment).
- High internal validity.
Cohort Study (Prospective)
- Moves Forward in time.
- Cause → Effect.
- Start with exposure (e.g., Smokers) and follow up to see if they develop disease (Lung Cancer).
Case-Control Study (Retrospective)
- Moves Backward in time.
- Effect → Cause.
- Start with Disease (e.g., Lung Cancer patients) and look back at history (Did they smoke?).
- Best for: Rare diseases.
6. Reliability vs. Validity
- Reliability (Consistency): If you measure the same thing twice, do you get the same result?
Example: If a goniometer reads 45° today and 45° tomorrow for the same joint position, it is reliable. - Validity (Accuracy): Does the tool measure what it is supposed to measure?
Example: Does the Lachman test actually detect an ACL tear? If yes, it is valid.
Practice Quiz: Research & Biostatistics (10 MCQs)
Click on the option you think is correct to see the answer and explanation.
Q1. Manual Muscle Testing (MMT) grading (0, 1, 2, 3, 4, 5) is an example of which scale of measurement?
Reason: MMT grades have a specific order (rank), but the difference between grade 3 and 4 is not mathematically equal to the difference between 1 and 2.
Q2. Which measure of central tendency is most appropriate for skewed data (data with outliers)?
Reason: The Mean is pulled towards extreme values (outliers). The Median represents the middle position and is resistant to outliers.
Q3. A researcher wants to compare the effectiveness of three different types of exercises (Isometric, Isotonic, Isokinetic) on quadriceps strength. Which statistical test should be used?
Reason: ANOVA is used when comparing the means of more than two groups. A t-test is limited to two groups.
Q4. Which of the following variables is an example of Ratio data?
Reason: Walking speed has a true absolute zero (0 m/s means no movement). Temperature in Celsius has an arbitrary zero.
Q5. A study that starts with patients who already have a disease and looks back into their history for risk factors is called:
Reason: Case-Control studies are retrospective (Effect → Cause). Cohort studies are prospective (Cause → Effect).
Q6. To analyze the association between two categorical variables (e.g., Smoking: Yes/No and Lung Cancer: Yes/No), which test is used?
Reason: Chi-Square is a non-parametric test used for categorical (nominal) data to check for associations.
Q7. Which sampling method involves dividing the population into subgroups based on characteristics (e.g., Age) and then randomly selecting from each?
Reason: This ensures that all subgroups (strata) are adequately represented in the final sample.
Q8. Reliability refers to:
Reason: Reliability means getting the same result upon repeated measurements. Accuracy refers to Validity.
Q9. What is the Type I error in hypothesis testing?
Reason: Also known as a "False Positive". You claim a difference exists (treatment works) when actually there is no difference.
Q10. Which is the highest level of evidence in the hierarchy of research designs?
Reason: While RCTs are the gold standard for *individual* experiments, a Systematic Review combines results from multiple RCTs, providing the highest level of evidence.
References
- Kothari, C. R. (2004). Research Methodology: Methods and Techniques. New Age International.
- Portney, L. G., & Watkins, M. P. (2009). Foundations of Clinical Research: Applications to Practice. Pearson/Prentice Hall.
- Mahajan, B. K. (2010). Methods in Biostatistics for Medical Students and Research Workers. Jaypee Brothers.
No comments:
Post a Comment