ETF2100 Chap.7 Interactions, Indicators and Qualitative Information

Interactions, Indicators and Qualitative Information

Establish the analytical object

Weeks 9 and 10 introduce qualitative information and interactions. An indicator shifts a fitted comparison relative to a chosen reference category; the zero category is not inherently normal or unimportant. An interaction changes the interpretation of the associated main terms because the effect of one variable now depends on another.

The safest reading reconstructs the fitted value for each relevant case instead of interpreting coefficients in isolation.

Labels must also track coding: reversing the indicator or changing the reference category changes coefficient signs and intercept meaning while leaving the fitted observations unchanged.

The chapter objective is to encode group differences and changing slopes without treating a reference category as a universal baseline. Begin by defining indicator variable at the scale used in the question.

Record whom or what indicator variable describes, its period or operating state, and evidence that distinguishes indicator variable from interaction. Without that discipline, indicator variable can quietly change meaning between the opening claim and the final recommendation.

Next, make reference category do explanatory work.

State the direction of reference category, the process it carries and the condition that keeps its link with indicator variable credible. A useful reference category note does not merely say that the relationship matters.

It identifies which observation establishes indicator variable, which observation tests reference category and which value of interaction would force a different account.

Use interaction as the chapter's discriminating lens. Compare at least two feasible cases and decide whether interaction strengthens, narrows or reverses the preferred result. If it cannot alter any conclusion, it is functioning as decoration.

Attach the comparison to the same unit, population or system boundary used for indicator variable and reference category.

Trace the operative relationship

A complete application of indicator variable has an actor, evidence, relationship and decision.

The actor has responsibility; evidence identifies the indicator variable state; reference category explains why action may work; and interaction supplies a review signal. This indicator variable–reference category–interaction structure makes ETF2100 reasoning auditable without turning one definition into a universal rule.

Consider a wage model with education, a training indicator and their interaction.

The education coefficient describes the fitted education slope when training equals zero. The training coefficient describes the fitted training contrast when education equals zero, which may be outside the observed or meaningful range. The interaction describes how the education slope differs for trained workers. Centre education at a meaningful value, such as 12 years, if the zero-point contrast is substantively useless.

Then compute fitted comparisons at several observed education values rather than declaring one permanent training effect.

Now change one condition: Change the training indicator's reference from untrained to trained. Predict which coefficients change sign or meaning and which fitted wage values remain invariant. Predict the direction of the result before consulting an example.

Explain whether the change affects the definition of indicator variable, the mechanism carried by reference category, the comparison represented by interaction, or only the confidence attached to the conclusion.

Keep the controlling limit visible: An interaction can describe heterogeneity in a conditional relationship; it does not by itself explain why the groups differ. This interaction limit is not ceremonial.

It specifies the observation, design feature or operating condition that separates a careful use of indicator variable from a claim that outruns reference category evidence.

Use the boundary as a test

For retrieval, close the explanation and reconstruct indicator variable, reference category and interaction in three different sentences: a definition, a relationship and a counter-case.

Then attach one concrete ETF2100 example to each. Reopen the interaction material only to correct the first missing indicator variable–reference category link; copying everything hides which analytical role failed.

For written or oral assessment, put the interaction conclusion after the reasoning.

Start with the requested decision, use indicator variable to establish the object and trace reference category before allowing interaction to challenge the preferred position. Report interaction at the scale earned by indicator variable evidence, preserving uncertainty and implementation constraints around reference category.

Create an error log specific to indicator variable.

Record the triggering fact, mistaken indicator variable inference, repaired relationship involving reference category, and evidence from interaction that distinguishes the two. Repeat the repaired reference category move on a different interaction case so feedback becomes a transferable diagnostic for indicator variable.

A strong final check asks four questions. Is indicator variable defined consistently?

Does reference category explain a process rather than repeat the outcome? Can interaction genuinely contradict the preferred answer? Does the last sentence remain inside this limit: An interaction can describe heterogeneity in a conditional relationship; it does not by itself explain why the groups differ.

If any indicator variable–reference category–interaction answer is no, revise that defective relationship rather than adding more description.

What this chapter covers

• 01

  indicator variable

• 02

  reference category

• 03

  interaction

• 04

  encode group differences and changing slopes without treating a reference category as a universal baseline

• 05

  An interaction can describe heterogeneity in a conditional relationship; it does not by itself explain why the groups differ.

Changed indicator variable case

Key terms

indicator variable

A variable that marks membership in a category, commonly with zero and one values.

reference category

The omitted or baseline category against which indicator coefficients are interpreted.

interaction

A model term allowing the association of one regressor to depend on another variable.

Interactions, Indicators and Qualitative Information FAQ

Exam move

Retrieve indicator variable, reference category and interaction; explain their relationship; apply them to the changed case; then test the result against the stated boundary.

Working through Interactions, Indicators and Qualitative Information in ETF2100? Sia is AskSia’s AI Econometrics tutor — ask any ETF2100 Interactions, Indicators and Qualitative Information question and get a clear, step-by-step explanation grounded in how ETF2100 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.