ZRH --:--:--

Questionnaires

2 min read9 sections
On this page

Definition:

  • In theory, not verifiable by external observation or records.
  • Has evaluative component
  • How strong agree or disagree
  • In contrast to beliefs (cognitive component): How healthy is a pizza?
  • In contrast to behavioral intentions: In the next 6 months, how likely is it that you will buy a car

Traditional View on Attitude

A psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor

Attitudes are Pre-existing: - Automatic activation of attitudes (retrievable: “file-drawer”) - stable and predictive of behaviors, resistant to persuasion

Problems of traditional view

In survey big differences are found depending on how question is asked.

Alternative View

“Making it up as you go along”

  • individuals do not typically possess true attitudes rather they construct opinion statements on the fly based on whatever considerations are momentarily salient

Schwarz 2007: evaluative judgments formed on the spot rather than as trait-like dispositions

Rasinksi’s way: Resovling these Divergent views

Attitudes vs. attitude expressions

  • Any single attitude measure will be an imperfect reflection of the underlying attitude
  • Attitudes are unobservable, global evaluations of objects

Some attitude expressions are construced on the spot, some are retrieved almost entirely intact from memory, others are a combination.

Schwarz Resolving

No attitudes, but contextualized construction of evaluations/judgments on the spot drawing on what is available to Respondent (R)

  • Stable attitudes = similar evalutions arising from R drawing on similar inputs or the some chronically accessible information.
  • Strong attitudes = R drawing on highly accessible information
  • Context effects = R drawing on contextual information when constructing evaluations
  • Nonattitude = Same number of conflicting information accessible to R at the same time

Context Effects

  • Subject of questionnaire
  • Interviewer
  • Interview setting
  • Instructions
  • Pictures that accompany the question
  • Wording of the question
  • Response options
  • Question order

Context Effects: Comprehension

  • Assimilation effects: Rs include prior question in interpreting/understanding the current item
  • Contrast effects: Rs exclude prior questions in interpreting and understanding the current item

Conetxt Effects: Retrieval

Context prompts the accessibility or retrieval of certain considerations (assimilation) or exclude the retrieval of certain considerations (contrast)

Context Effects: Judgement

The inclusion/exclusion model Constructing targets and standards

Specific vs. General Evaluations

Only ask specific question and if general question then asked them first.

Agree Disagree Scales (should be avoided)

Advantages

Fast for R Fewer “don’t know” R like these scales

Disadvantages

Acquiesence Bias - R wants to be polite and agreeable - R might defer to higher status (yes sure) - R adopts a satisfying strategy (do just enough)

Multi-barrled - multi information in statement

Disagree end is ambigious

How certain are you of your views on abortion? - Very certain, somewhat certain, not very certain, not certain at all

Don’t Know Option?

Back to top

Esc

Type at least two letters. Search covers every project and knowledge-base note.

oz@zrh: ~
QUICK DOSSIER · 30 SEC READ

OLIVER ZINGG

Psychologist and data scientist in Zürich. I design the study, model the data and explain what it means.

STATUS
Open to full-time data science roles from January 2027
NOW
Working student, data science & project management at Zurich Insurance (since Nov 2025)
THESIS
M.Sc. thesis (HSLU): leveraging double machine learning for robust decision-making, due December 2026
EDUCATION
M.Sc. Psychology (UZH, 5.6) · M.Sc. Applied Information & Data Science (HSLU, Dec 2026)
TOOLS
R, Python, SQL, Matlab · Power BI, Tableau · mixed models, ML, BERTopic

TOP CASE FILES

  1. Learning and memory formation in children and adolescents with psychiatric disordersMaster's thesis, UZH · 2023
  2. Support vector machines vs. CNNs for Alzheimer's disease classificationBachelor's thesis, UZH · 2021
  3. Simpson's paradox and mixed-effects modelsStatistics · 2023