Quantitative, Positivist Research (QtPR) involves using numeric data to study how people and digital technologies interact. It typically assumes there is an objective reality—that if measured correctly, we can gather valid insights about how things work.

---

Focus on Numbers

In quantitative research, we work primarily with data in numeric form. Examples include:

• Survey responses on a 1–5 or 1–10 scale.
• Clicks on a website or usage logs in an app.
• Sales figures from archived company data.

Why Numbers Matter

Researchers use numeric scales to represent otherwise subjective ideas. For instance, a 1–5 scale might measure website satisfaction (1 = “Not satisfied at all,” 5 = “Very satisfied”). Even though satisfaction is a mental or emotional concept, reducing it to numbers helps in statistical analysis.

Statistical Methods: Common, Not Mandatory

Quantitative studies often use statistical tools, but the essential point is that the data are numeric. Even something as simple as counting the number of positive vs. negative comments (and comparing the totals) can qualify as quantitative research.

---

How We Gather Numeric Data

1. Archival Data
   • Example: Analyzing past sales records to see whether implementing a new information system boosted revenue.
2. Structured Questionnaires or Surveys
   • Example: Sending a survey to 1,000 customers, asking them to rate their experience from 1 (poor) to 10 (excellent).
3. Coding Interviews or Web Posts
   • Example: Assigning numeric codes to statements in interviews (e.g., “user-friendly interface” = 1, “confusing interface” = 0).
4. Transactional Data
   • Example: Counting how many times users log into an app, how long they stay, or how many items they purchase online.

Interpreting the Numbers

Once the data are collected, researchers analyze them to extract meaning:

• If 80% of users log in every day, it suggests high adoption.
• If survey responses average 4.7 out of 5 for “ease of use,” we infer the system is user-friendly.

Simple Example

Imagine running a small online store. You ask customers (1–5) how satisfied they are, and you also track monthly returns (how many customers come back). Both are numeric measures. You might compare the data before and after adding a new site feature to see if satisfaction and loyalty improved.

---

2. (Post-) Positivist Philosophy

QtPR is positivist in its assumption that an objective reality exists, but is more accurately post-positivist, acknowledging that we can never measure reality perfectly.

Positivist vs. Post-Positivist

• Positivist: Believes an objective reality exists and that researchers can uncover the truth about how things work if they apply the right methods.
• Post-Positivist: Similar to positivism, yet recognizes that all observations and measurements have some degree of imperfection. We still aim for objectivity, but accept that our knowledge is never 100% certain.

Contrast with Interpretive Research

• Interpretive scholars see reality as socially constructed (through language, culture, and shared meanings).
• They focus less on numeric measurement and more on understanding how people interpret their experiences.

Common Ground

Both positivist/post-positivist and interpretive researchers agree that constructs like “customer satisfaction” are human concepts used to make sense of the world—even if they differ on whether these concepts can be measured objectively.

Everyday Example

• Positivist/Post-Positivist Approach: “How many hours do people spend on social media, and how does that correlate with their anxiety levels?” (Measured via app usage data and a standardized anxiety questionnaire.)
• Interpretive Approach: “Let’s do in-depth interviews about how people feel when using social media, and find common themes in their narratives.”

---

3. Cornerstones of QtPR

Quantitative Data Focus

QtPR relies on numeric evidence—survey ratings, system logs, coded interviews—to measure concepts. While sophisticated statistical techniques are common, they are not required so long as the data remain fundamentally numeric.

(Post-) Positivist View

Reality is assumed to be “out there” and partially knowable. However, post-positivists admit we can’t capture it perfectly. This differs from interpretive research, which suggests reality is shaped mainly by social context and language.

Why Combine the Two Cornerstones?

• The numeric approach provides a structured, rigorous method for studying cause-and-effect relationships.
• The post-positivist stance offers realism about the limits of measurement.

QtPR is especially useful when you want to see if changing one variable (e.g., adding a chatbot) impacts another (e.g., customer satisfaction scores) in a measurable way.

---

4. Understanding “Theory” in QtPR

In QtPR, a theory often takes the form of lawlike statements that explain how or why certain factors cause changes in others.

Example:

“In an online store, making the website more user-friendly (Factor A) leads to higher customer satisfaction (Factor B).”

Why Call It “Lawlike”?

Because it describes a consistent, predictable relationship: improving website design should consistently correlate with an uptick in satisfaction, implying causality.

Gregor’s Definition of Theory (2006)

Gregor describes theory as explaining:

1. What Is: The main variables or factors.
2. How: The relationships among them.
3. Why: The underlying reasons or mechanisms.
4. When/Where: The contexts or conditions that influence these relationships.
5. What Will Be: Predictions or outcomes we expect to see.

A strong theory identifies which factors matter and clarifies how, why, and under what conditions they matter, as well as what outcomes we can predict.

Practical Example

Researching how “training hours” (Factor A) affect “employee performance” (Factor B):

• What Is: Training hours and employee performance are your constructs.
• How: More training hours might increase performance.
• Why: Better training boosts skills, reduces errors, and improves speed.
• When/Where: This effect could be strongest in first-time system adopters (e.g., a startup).
• What Will Be: Raising training from 2 to 5 hours a week should yield a measurable improvement in performance test scores.

Such a statement is a testable proposition in QtPR, as it implies Factor A causes Factor B to change.

---

5. How QtPR Is Used

QtPR can be employed to generate new theories or to evaluate existing theories.

1. Theory Generation

Steps

1. Identify Constructs: Pinpoint the big ideas (e.g., user satisfaction, system quality, training hours).
2. Build Measurements: Decide how to quantify these constructs (scales, logs, etc.).
3. Link Them: Articulate the relationships (“If system quality increases, then user satisfaction also goes up.”).

Example

You notice that employee morale (Construct A) might connect to use of a new HR system (Construct B). You design a 1–5 “work happiness” survey to measure morale and track each employee’s system usage. Then you hypothesize: “Higher usage hours correlate with higher morale.” You’ve proposed a new theory about how system usage might drive morale.

2. Theory Evaluation

Steps

1. Collect Data: Gather numeric info about constructs mentioned in an existing theory.
2. Test Relationships: Use statistical or data-modeling methods to see if predicted relationships hold true.
3. Assess Fit: If data match the theory, it’s supported (but not proven forever); if not, refine or reject it.

Example

A published theory says: “Employee satisfaction depends on system reliability and supervisor support.” You collect:

• Reliability: Downtime per month.
• Supervisor Support: A 1–5 rating of how supportive managers are.
• Employee Satisfaction: Another 1–5 scale.

Plug these into a statistical model. If higher reliability and stronger supervisor support indeed lead to higher satisfaction, the theory holds—for now.

---

6. The Hypothetico-Deductive Model

This model underlies much of theory evaluation in QtPR and is a classic view of how science works.

1. Develop Hypotheses: “X causes Y.”
2. Design a Study: Plan how you’ll test X and Y.
3. Collect Data: Gather numerical evidence (e.g., surveys, usage logs).
4. Analyze Data: Check if results align or contradict your hypothesis.
5. Interpret and Report: Draw conclusions and share findings.

Why “Hypothetico-Deductive”?

• You start with a hypothesis (an educated guess).
• You deduce what outcomes you’d expect if the hypothesis is correct.
• You test whether real-world data match those predicted outcomes.

---

7. Objective Reality—But Not Perfect Measurement

Traditional Positivist View

Positivists historically believed the world is objectively real and can be precisely measured with the right tools.

Modern Twist: Post-Positivism

Post-positivists acknowledge that all measurement has some error. Even measuring temperature depends on theories about how thermometers or digital sensors react to heat. Perfect objectivity is unattainable, though we try our best.

Everyday Example

When a digital thermometer reads 98.6°F, it looks objective, but remember there’s a sensor interpreting electrical signals based on theory about how heat impacts electronics. There could be a small margin of error.

---

8. Falsification vs. Confirmation

Karl Popper’s Falsification

Popper proposed that science advances by trying to prove theories wrong. One contradictory instance can disprove an entire theory. Example: If your theory is “All swans are white,” finding one black swan falsifies it.

The Reality: Confirmation

In the social sciences, many researchers instead look for confirmation of their hypotheses. Strict falsification is rarely done because a single contradictory finding might be caused by poor data or small samples. Researchers often set up an alternative hypothesis (“A is positively related to B”) rather than a null hypothesis of “no effect,” making it easier to demonstrate confirmation.

---

9. The Critical-Realist Perspective

Cook and Campbell (1979) argue for a critical-realist view:

1. A real world exists.
2. We can never measure it perfectly, even with advanced methods.
3. We must remain critical about the conclusions we draw and avoid claiming we’ve found an absolute, unchanging truth.

---

10. Quick Example in Action

1. Theory: “A faster-loading website increases user satisfaction.”
2. Hypothesis: “Reducing page load times by 50% raises satisfaction by 10 points on a 100-point scale.”
3. Design & Data: Compare two groups—one with normal load times, one with optimized load times—and measure each group’s satisfaction.
4. Analyze: Check if the optimized group’s satisfaction is indeed 10 points higher.
5. Interpret: If yes, the hypothesis is supported (confirmed). If not, we’ve “failed to support” or potentially falsified the hypothesis.
6. Refine: Perhaps speed improvements only matter for certain user segments, leading you to revise the theory.

This encapsulates how QtPR works: relying on quantitative data, assuming a real phenomenon can be observed (though imperfectly), and testing theories using the hypothetico-deductive approach—seeking both confirmation and potential falsification.

---

In Summary

1. QtPR (Quantitative, Post-Positivist Research) is about using numbers and statistics to investigate questions about how people, organizations, and technology interact.
2. It can create new theories or evaluate existing theories, following the hypothetico-deductive model (hypothesis → data collection → analysis → interpretation).
3. Originally positivist, it has evolved into post-positivism, accepting that no measurement is completely perfect.
4. Falsification (Popper’s idea) is an ideal goal, but in practice, researchers often look for confirmation or support of their hypotheses.
5. Over time, these cycles of data-based testing refine our theories, inching us closer to an understanding of “how things really work,” while recognizing we can’t achieve total certainty.

Final Takeaway

QtPR:

• Relies on numeric data and systematic methods.
• Aims for objectivity, while acknowledging measurement limitations.
• Operates cyclically: form hypotheses, collect data, analyze, interpret, and refine.
• Falls under post-positivism, embracing the complexity of reality and the imperfection of our measures.