**QUANTITATIVE RESEARCH TERMS EXPLAINED IN DETAIL (WITH 3-SENTENCE DEFINITIONS AND

EXAMPLES)**

---

### 1. Population and Sample

* **Population**: The population is the complete set of individuals or elements the researcher
wants to study. It represents the whole group about which data is needed. *Example*: All university
students in Morocco.

* **Sample**: A sample is a subset of the population selected for actual study. It should be
representative to allow generalization of findings. *Example*: 500 students randomly selected from
10 Moroccan universities.

### 2. Sampling Frame

* A sampling frame is the actual list from which a sample is drawn. It must accurately reflect the
target population to avoid bias. *Example*: A university's official student enrollment list.

### 3. Random Errors

* Random errors are unpredictable variations in data due to chance. They affect reliability but not
validity. *Example*: A respondent accidentally misreads a survey question.

### 4. Systematic Errors

* Systematic errors occur due to consistent issues in research design or tools. They introduce bias
and affect validity. *Example*: A leading survey question that influences responses.

### 5. Coverage Error

* Coverage error arises when the sampling frame doesn’t include all segments of the population. It
leads to biased results because some individuals have no chance of being selected. *Example*:
Students without internet access are excluded from an online survey.
### 6. Nonresponse Error

* Nonresponse error occurs when selected individuals don’t participate. It reduces

### 7. Self-selection Error

* Self-selection error happens when individuals decide themselves whether to participate. This can
lead to a biased sample if certain types of people are more likely to respond. *Example*: Only
students interested in the survey topic participate.

---

### 8. Types of Sampling

* **Probability Sampling**: Every individual in the population has a known, non-zero chance of
being selected. It allows for generalizing results to the whole population. *Example*: Random digit
dialing.

* **Cluster Sampling**: The population is divided into clusters, and entire clusters are randomly
selected. It’s used when it’s difficult to create a list of the whole population. *Example*: Randomly
selecting 5 universities, then surveying all students there.

* **Stratified Sampling**: The population is divided into subgroups (strata), and samples are taken
from each. It ensures all subgroups are represented. *Example*: Sampling 100 students from each
academic year.

* **Purposive Sampling**: Specific participants are chosen because they meet certain criteria. It is
used in exploratory research. *Example*: Only surveying final-year engineering students.

* **Quota Sampling**: Researchers ensure specific traits or proportions in the sample. It resembles
stratified sampling but selection isn’t random. *Example*: Selecting 50% male and 50% female
respondents.
* **Snowball Sampling**: Existing participants recruit others. It’s useful for hard-to-reach
populations. *Example*: Surveying influencers who invite others to participate.

* **Convenience Sampling**: Participants are selected based on availability. It’s easy and quick but
less representative. *Example*: Surveying classmates because they are nearby.

---

### 9. Probability Theory and Statistical Concepts

* **Probability Theory**: Forms the basis of inferential statistics, helping researchers predict the
likelihood of outcomes. It is used to make generalizations from a sample. *Example*: Calculating the
chance that survey results represent the population.

* **Law of Large Numbers**: States that as sample size increases, the sample mean gets closer to
the population mean. Larger samples produce more reliable results. *Example*: A survey of 5,000
people gives a more accurate average than one of 50.

* **Central Limit Theorem**: The distribution of the sample mean becomes normal as sample size
increases. This allows researchers to use normal distribution in hypothesis testing. *Example*:
Averages of student grades from repeated samples follow a normal distribution.

* **Confidence Level**: Indicates how sure we are that a sample result reflects the population. A
common confidence level is 95%. *Example*: We are 95% confident the average student GPA lies
between 2.8 and 3.2.

* **Null Hypothesis (H0)**: Assumes no effect or relationship exists in the population. It's tested to
determine statistical significance. *Example*: There's no GPA difference between male and female
students.

* **p-value**: The probability that the observed results happened by chance. A p-value below 0.05
typically indicates statistical significance. *Example*: A p-value of 0.01 suggests a strong likelihood
the result isn’t random.

* **Statistical vs Practical Significance**: Statistical significance shows results are unlikely due to
chance; practical significance means they matter in real life. Both should be considered when
interpreting results. *Example*: A small GPA increase may be statistically significant but not
practically useful.

---

### 10. Descriptive and Inferential Statistics

* **Mode**: The most frequently occurring value in a dataset. It gives insight into common
responses. *Example*: Mode of student ages in a class is 20.

* **Median**: The middle value when data is ordered. It’s useful when data has outliers. *Example*:
Median household income avoids the effect of very high incomes.

* **Mean**: The average of all values. It’s widely used but affected by outliers. *Example*: The
mean score of students on a test is 75%.

* **Range**: The difference between the highest and lowest values. It shows data spread.
*Example*: Test scores ranging from 40 to 95 have a range of 55.

* **Standard Deviation**: Measures how spread out values are from the mean. A high standard
deviation means more variability. *Example*: A low SD in test scores means most students scored
similarly.

* **Bivariate Statistics**: Examine the relationship between two variables. Used in correlation and
regression. *Example*: Analyzing the link between study hours and GPA.

* **Correlation Analysis**: Measures the strength and direction of a relationship. A value near +1 or
-1 shows a strong correlation. *Example*: A correlation of 0.8 between sleep and performance
indicates a strong positive link.

* **Regression Analysis**: Predicts the value of a dependent variable based on one or more
independent variables. It shows the strength and type of relationships. *Example*: Using study hours
to predict exam scores.

* **Multiple Regression**: Involves more than one independent variable. It identifies the effect of
each predictor. *Example*: Predicting GPA using study hours, class attendance, and sleep.
* **Non-parametric Tests**: Used when data doesn’t meet assumptions for parametric tests, like
normal distribution. They’re more flexible but less powerful. *Example*: Mann-Whitney U test for
comparing two groups without assuming normality.

---

### 11. Experimental Design

* **Random Assignment**: Participants are randomly placed into experimental or control groups. It
ensures comparability and reduces bias. *Example*: Assigning students to two groups using a
random number generator.

* **Treatment and Control Groups**: Treatment group receives the intervention; control group does
not. Helps in evaluating effects. *Example*: One group gets a study app, another does not.

* **Observation Bias**: Participants change behavior because they know they are being watched. It
affects data validity. *Example*: Students perform better under observation during an experiment.

* **Social Desirability Bias**: Respondents give answers that make them look good. It compromises
honesty. *Example*: Saying they study more than they actually do.

* **Response Bias**: Tendency to answer inaccurately or falsely. It can stem from wording or
misunderstanding. *Example*: Misinterpreting a question and answering incorrectly.

* **A/B Testing**: Comparing two variants to see which performs better. Widely used in digital
marketing. *Example*: Testing two versions of a webpage to see which gets more clicks.

* **Digital Field Experiments**: Conducted online in real-world settings. They help test behavior in
natural digital environments. *Example*: Testing a new app feature with live users.

* **Physical Field Experiments**: Take place in real offline environments. They help study real
behavior under natural conditions. *Example*: Observing classroom behavior after introducing new
teaching methods.
* **Quasi-Experiments**: Lack random assignment but still compare groups. Used when
randomization isn’t possible. *Example*: Comparing two existing school classes with different
teaching styles.

* **Usability Tests**: Assess how easily users can use a product or system. Common in software
testing. *Example*: Asking users to complete tasks on a new app to evaluate ease of use.

---

### 12. Digital Research Methods

* **Commercial Databases**: Collections of data sold by companies for research use. They save time
but may be costly. *Example*: Buying a consumer behavior dataset from Nielsen.

* **Scraping Data**: Automatically extracting data from websites. Useful for gathering large-scale
online information. *Example*: Scraping Twitter data for sentiment analysis.

* **APIs**: Allow researchers to access structured data from digital platforms. APIs ensure legal and
efficient data collection. *Example*: Using the YouTube API to get video statistics.

* **Digital Trace Data**: Data left behind by users online, often passively. It reflects real behavior
rather than self-reporting. *Example*: Analyzing website click patterns.

* **Integrating Methods**: Combining qualitative and quantitative approaches. Enhances depth and
reliability. *Example*: Using interviews and surveys in the same study.

* **Enriching Methods**: One method deepens the findings of another. It provides context or
explanation. *Example*: Using interviews to understand surprising survey results.

* **Contesting Methods**: Using different methods to challenge each other’s findings. Ensures rigor
and checks biases. *Example*: Comparing content analysis and survey results to detect
contradictions.

---
### 13. Surveys and Question Design

* **Self-administered Surveys**: Completed by the respondent without researcher presence. They

* **Researcher-administered Surveys**: Conducted face-to-face or by phone. Allow clarification of

* **Interviewer Effect**: The interviewer’s presence influences responses. This can lead to biased
answers. *Example*: A respondent gives more positive answers in person than online.

* **Behavioral Questions**: Ask about what people do or have done. They give insight into actual
behavior. *Example*: How many hours do you study each week?

* **Attitudinal Questions**: Explore opinions, beliefs, or feelings. They measure attitudes toward
topics. *Example*: Do you agree that online learning is effective?

* **Knowledge Questions**: Test what respondents know about a topic. Useful in evaluating
awareness or information levels. *Example*: What is the capital of Canada?

* **Vignettes**: Present hypothetical scenarios to assess judgments. They help explore complex
decision-making. *Example*: A short story about a cheating student to see how participants react.

* **Closed-ended Questions**: Provide predefined answer choices. Easy to analyze quantitatively.

* **Open-ended Questions**: Allow respondents to answer freely. Provide rich, qualitative data.
*Example*: What do you think about online education?

---

### 14. Study Designs

* **Cross-sectional Research**: Data is collected at one point in time. It’s useful for capturing a
snapshot. *Example*: A survey on student stress during final exams.
* **Longitudinal Research**: Data collected over time to observe changes. Helps study trends and
developments. *Example*: Following students from freshman to senior year.

* **Panel Studies**: Involve repeated data collection from the same individuals. Useful for tracking
personal change. *Example*: Surveying the same group of teachers every year.

* **Cohort Studies**: Follow a group with a shared characteristic. Helps examine how experiences
affect them over time. *Example*: Studying students who enrolled in 2020.

---

### 15. Data Structure

* **Cases**: Individual units of analysis, such as people or organizations. Each case provides one set
of responses. *Example*: Each student in a survey is a case.

* **Variables**: Characteristics or attributes measured in a study. Variables vary between cases.

* **Observations**: Specific values recorded for each variable. They form the dataset. *Example*: A
student's GPA of 3.5.

* **Content Analysis**: Systematic coding and counting of content. Used to analyze media or texts
quantitatively. *Example*: Counting the number of times “climate change” appears in news articles.