# Analysing Social Processes

Sociology I · Contemporary Social Processes · https://tryals.app/learn/sociology-i/analysing-social-processes

## Measuring Something That Moves

A social process is a patterned change over time. Studying one requires turning abstract concepts into measurable data. This process is called **operationalisation**.

Two properties of any measure must be kept distinct:
- **Reliability**: consistency across repeated measurements.
- **Validity**: whether the tool measures what it claims to measure.

A faulty scale can be reliable yet invalid. Sociological measures usually struggle with validity because concepts lack obvious countable units.

| Technique | Answers well | Answers badly |
|---|---|---|
| Probability survey | How common is X in a population | What X means to those who do it |
| In-depth interview | How people account for their conduct | How representative that account is |
| Ethnography | What actually happens | Population estimates |
| Documentary analysis | Change over long periods | Unrecorded questions |
| Experiment | Whether A causes B | Real-world generalisability |

## Sampling and Indicators

Sample design matters more than raw sample size.

In 1936, the **Literary Digest** polled **2.4 million** people (2.4 million returned ballots). It wrongly predicted victory for Alf Landon over Roosevelt. George Gallup polled 50 thousand and predicted correctly. A biased sample yields a precise estimate of the wrong quantity.

Watch for two methodological traps:
- **ecological fallacy**: inferring individual traits from aggregate data (Robinson, 1950).
- **reactivity**: subjects altering behaviour when observed, first noted at Western Electric.

**Indicators** track wider social change:
- **Human Development Index**: tracks income, health, and schooling (since 1990).
- **Gini coefficient**: summarises inequality from 0 to 1.
- **at-risk-of-poverty**: measures relative poverty rates.

> **Common pitfall:** treating sample size as a guarantee of quality. Size reduces random error, but it never reduces systematic bias.

## Practice questions

10 of this lesson's 15 practice questions, with answers. The full set is in the app.

### 1. The Literary Digest received about 2.4 million returned ballots in 1936 and predicted the wrong winner, while Gallup was right with roughly fifty thousand cases. Why did the far larger sample perform worse?

A. Phone and car lists plus voluntary returns made the sample unlike the electorate
B. High return volume caused aggregation bias and ecological fallacies about voters
C. Reactivity bias arose when participants changed declared votes after the ballot
D. Probability formulas failed at scale, adding random errors to projected outcomes

**Answer:** A. Phone and car lists plus voluntary returns made the sample unlike the electorate

**Why:** Two distortions compounded: the frame excluded poorer households in the middle of the Depression, and returning a mailed ballot was itself voluntary, which selected the motivated. Neither error shrinks as the sample grows, so the extra millions bought precision about a population that was not the electorate.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/the-literary-digest-received-about-2-4-million-returned-ballots-in

### 2. Match each research technique to the question it answers best.

**Answer:**

- Probability survey → How common something is across a whole population
- In-depth interview → How people themselves account for what they do
- Ethnography → What happens in a setting, as against what is reported about it
- Documentary and secondary analysis → How a pattern has changed across decades already recorded

**Why:** The list is not a hierarchy. Each technique is strong exactly where another is weak, which is why mixed designs are common: a survey establishes how widespread something is and interviews establish what those doing it take themselves to be doing, and neither substitutes for the other.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/match-each-research-technique-to-the-question-it-answers-best

### 3. A measure can be highly reliable and still fail completely as a measure of the concept it names.

**Answer:** True

**Why:** **True**, and this is the more common failure in sociology. Repeating a badly worded question gives the same distorted answer every time, which looks like a well-behaved instrument. Reliability is easy to demonstrate and validity is hard, so published measures skew towards the first.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/a-measure-can-be-highly-reliable-and-still-fail-completely-as-a

### 4. Sort each design decision by whether it threatens the reliability or the validity of a measure.

**Answer:**

- Threatens reliability: Different interviewers record the same answer in different categories, The coding scheme is applied inconsistently across the fieldwork period
- Threatens validity: Church attendance is used as the sole indicator of religious belief, Examination results are used to measure the quality of a school

**Why:** The last item is the one worth arguing over. Examination results are highly reliable and measure the intake of a school at least as much as its teaching, which is why value-added measures were invented, an attempt to rescue validity from an instrument whose reliability was never in doubt.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/sort-each-design-decision-by-whether-it-threatens-the-reliability-or

### 5. Which statements about the ecological fallacy are correct?

A. It shows that aggregate data are useless for any sociological purpose
B. It is the error of inferring something about individuals from data aggregated over areas
C. A positive correlation between two variables across districts can coexist with a negative one between the same variables across individuals
D. It was named by W. S. Robinson in 1950 from an analysis of literacy and nativity in the United States

**Answer:** B. It is the error of inferring something about individuals from data aggregated over areas; C. A positive correlation between two variables across districts can coexist with a negative one between the same variables across individuals; D. It was named by W. S. Robinson in 1950 from an analysis of literacy and nativity in the United States

**Why:** Robinson’s original example had literacy correlating with the foreign-born share across states while foreign-born individuals were on average less literate, because immigrants settled in states with better schools. The fallacy is about the level of inference, not about the quality of the data: district-level questions can be answered from district-level data without any error at all.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/which-statements-about-the-ecological-fallacy-are-correct

### 6. Arrange the steps of operationalising a sociological concept, in order.

**Answer:**

1. State the concept the research is about
2. Specify the dimensions the concept is taken to have
3. Choose observable indicators for each dimension
4. Decide how the indicators will be recorded and coded
5. Test whether the resulting measure behaves as the concept should

**Why:** Skipping the second step is the usual failure. A researcher who moves straight from "social capital" to a question about club membership has silently decided that social capital has one dimension and that it is associational, and every later finding inherits that decision without ever stating it.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/arrange-the-steps-of-operationalising-a-sociological-concept-in

### 7. The Human Development Index has been published annually by the United Nations Development Programme since which year?

**Answer:** 1990

**Why:** **1990.** It was designed by Mahbub ul Haq with Amartya Sen as an explicit rebuke to income-only rankings, combining life expectancy, schooling and income. Its weaknesses are the price of that simplicity: three dimensions, equally weighted by fiat, and nothing about distribution.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/the-human-development-index-has-been-published-annually-by-the-united

### 8. A researcher chooses in-depth interviews over a probability survey to study why employees resist corporate restructuring. What trade-off follows from this methodological choice?

A. She isolates causal mechanisms but loses real-world generalisability.
B. She eliminates subject reactivity but loses historical depth across time.
C. She captures subjective meanings but loses population representativeness.
D. She establishes high reliability but loses validity in concept metrics.

**Answer:** C. She captures subjective meanings but loses population representativeness.

**Why:** Qualitative accounts clarify how participants make sense of events, whereas statistical surveys determine prevalence across whole populations. Choosing interviews privileges interpretative depth over representativeness, not experimental control or reactivity removal.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/a-researcher-chooses-in-depth-interviews-over-a-probability-survey-to

### 9. Increasing the size of a self-selected sample reduces the bias in its estimates.

**Answer:** False

**Why:** **False.** Size reduces sampling error, the random component. Bias is fixed by the selection procedure and does not shrink at all; a larger self-selected sample simply estimates the wrong quantity more precisely, which is the trap that destroyed the Literary Digest.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/increasing-the-size-of-a-self-selected-sample-reduces-the-bias-in-its

### 10. The Gini coefficient reduces an entire income distribution to a single number between zero and one. What does a sociologist give up in accepting that summary?

A. Details on where inequality sits, since distinct distributions can share a score
B. The ability to track decadal trends, since the index measures only static wealth
C. Validity needed to avoid ecological fallacy when linking macro data to individuals
D. Standardised data for cross-country comparison, as survey income definitions vary

**Answer:** A. Details on where inequality sits, since distinct distributions can share a score

**Why:** A single statistic cannot distinguish a society with a detached elite from one with an excluded bottom, and the two call for entirely different policies. That is why Gini coefficients are normally reported alongside decile ratios or top income shares, which say where the distance is.

Page: https://tryals.app/practice/sociology-i/analysing-social-processes/the-gini-coefficient-reduces-an-entire-income-distribution-to-a