We’ve decided we need to test SuperPaste, and we know how many children to pick (Sample Size), but we must be very careful about who we pick! If we select the wrong children, the study’s results won’t apply to the larger world, making the whole effort useless.
This chapter is about Selection of Study Population—the art of choosing the right children for our study.
1. 🔍 The Journey from Population to Sample
Choosing the study group is a step-by-step process of narrowing down a large, general group to the specific heroes we need.
| Terminology | Simple Idea | Defining Criteria |
| Population | The largest group (e.g., The Population of India). | The entire geographical area. |
| Target Population | The large group we want the results to apply to. | Clinical and Demographic characteristics. |
| Accessible Population | The group we can actually reach and study (a subset of the Target Population). | Geographical and Temporal (time) characteristics. |
| Study Sample | The small, final group of children selected from the Accessible Population. | Defined by Sampling Procedure and Exclusion Criteria. |
Example using SuperPaste:
- Target Population: All school-going children in India (Demographic and Clinical focus).
- Accessible Population: All children attending two specific schools in Chennai, recruited between January 1 and December 31, 2025 (Geographical and Temporal focus).
2. ⚖️ The Balance of Validity
When we choose our subjects, we have to balance two types of “truth”:
A. Internal Validity (The Truth Inside the Study)
- Simple Idea: This is the degree to which our independent variable (SuperPaste) truly caused the change in the dependent variable (fewer Dental Caries) within our study group.
- Goal: To have high Internal Validity, our study must be perfect, consistent, and free from error.
B. External Validity (Generalizability)
- Simple Idea: This is the degree to which the findings from our small study can be generalized or applied to the larger Target Population (all school-going children).
- Trade-Off: Studies with high Internal Validity (very specific, controlled subjects) are sometimes harder to generalize because the subjects were too perfect or unique. We must find a trade-off between the two.
- Example: Findings on the strength of association (like hypertension being a risk factor) are often generalizable (External Validity is good), but findings on a specific Prevalence (how common hypertension is) are often not generalizable (specific to that study area).
3. 📝 The Gatekeepers: Selection Criteria
We use two types of criteria to define exactly who gets into our Study Sample:
A. Inclusion Criteria
These are the main characteristics that define the Target Population and the Accessible Population. They are the necessary traits relevant to the research question.
- Clinical: Subjects must be free of chronic gum disease.
- Demographic: Must be between 6 and 10 years of age.
- Geographical/Temporal: Must attend School A or B during the recruitment period.
B. Exclusion Criteria
These are specific reasons a subject must be avoided because they might harm the study’s quality, follow-up, or ethical integrity.
| Exclusion Goal | Simple Reason to Exclude | SuperPaste Example |
| Interfere with Follow-up | High chance of moving out of the area or dropping out. | Children whose families are planning to move in the next 6 months. |
| Interfere with Data Quality | Subjects already using a similar intervention. | Children who already use a prescription-strength fluoride rinse. |
| Ethical/Safety Concerns | High risk of possible adverse effects from the intervention. | Children with a known hypersensitivity (allergy) to the ingredients in SuperPaste. |
4. 🚀 Recruitment Strategies and Feasibility
Feasibility (Can we actually do this?) is the most important factor in choosing our source of subjects.
A. Sources of Population
| Source | Description | Best for… | Challenge |
| Clinical Population | Patients attending hospitals or clinics. | Studies involving patients with a specific condition (e.g., advanced Dental Caries). | May not represent the average person (low Generalizability). |
| Community Population | Subjects selected directly from the general public (house-to-house or schools). | Studies on healthy subjects (e.g., Vaccine Efficacy Studies). | Difficult, time-consuming, and expensive to recruit. |
B. Dealing with Non-Response
Non-response (subjects refusing to join or dropping out, called Loss to Follow-up) severely harms Internal Validity and Generalizability.
To tackle this, researchers:
- Make Repeated Contact Attempts for subjects who were initially unavailable.
- Reduce Discomfort by ensuring study instruments are not invasive, using bilingual staff, and providing translated questionnaires to improve acceptability.
- Offer Incentives (where ethically appropriate) to encourage participants to adhere to the study.
By carefully applying the Inclusion and Exclusion Criteria and managing non-response, we ensure we recruit an adequate size of subjects that are representative of the whole population.
Leave a Reply