Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

Secular Trend Confounding

Also Known As: Time trend confounding Historical trend bias
Aspect ID: secular_trend_confounding

Definition

Secular trend confounding occurs when long-term background trends in both the exposure and outcome variables create the appearance of a causal relationship between them. If mobile phone ownership and diabetes prevalence both increase over two decades due to broader economic development, they will appear correlated even if mobile phones have no effect on diabetes.

Examples

An analysis of national data shows that organic food sales and autism diagnoses both rose dramatically from 2000 to 2015. Correlation = 0.99. Both trends are real, but both are independently driven by increased awareness, income, and screening — not by a causal link.

A wellness blogger plots smartphone ownership rates against childhood obesity rates from 2005 to 2020 and finds near-perfect correlation. Both trends are real, but both are independently driven by broad socioeconomic and lifestyle shifts over the same decade. Phones did not cause obesity — they simply rose in parallel.

A financial analyst reports that the number of women in the workforce correlates with rising household debt levels over a 30-year period. Both trends are genuine, but both are independently driven by long-term economic pressures, inflation, and changing social norms — not by any causal link between female employment and borrowing.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Do both the exposure and outcome variables show long-term trends over the study period?

    Type: binary
  2. 2

    Were the trends adjusted for in the analysis (e.g., by detrending or including time as a covariate)?

    Type: binary
  3. 3

    Could changes in both variables be independently driven by broader societal or economic trends?

    Type: binary
  4. 4

    Does the apparent relationship between exposure and outcome persist when the time trend is controlled?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.