Project 04: Who Believes in Astrology, and Where?
Source: bigastrologybook.com/2/research/19/project-4 Archive Date: 2026-03-21 Book: The Big Astrology Book of Research by Renay Oshop Datasets: Pew Research Center (N=4,729) + Google Trends state-level interest + US Census demographics
Research Question
What geographic, demographic, and cultural factors predict interest in and belief in astrology across the United States? Can the variation between states be explained by measurable demographic variables?
Hypothesis
Astrological interest varies systematically by geography and demographics — and those patterns, once mapped, reveal something meaningful about the cultural function astrology serves in contemporary America.
A Note on Two Different Measures
This project uses two distinct data sources that measure two different things, and it's worth being precise about them upfront:
- Google Trends "horoscope" search interest — measures engagement and curiosity. Someone searching for their horoscope is not necessarily a believer; they might be a skeptic, a casual reader, or just bored. This is a behavioral measure.
- Pew Research Center survey data — measures stated belief in astrology. This is a conviction measure, from a proper nationally representative survey (N=4,729, 2017–2018).
Both datasets tell the same demographic story, but they're not interchangeable. The findings below draw on both and note which is which.
Data transparency note: The Google Trends state-level figures are approximate — the analysis used cached 2023 data rather than live API output. The Pew survey figures are published and fully verified.
Data
| Source | Description |
|---|---|
| Pew Research Center (2017–2018) | National survey, N=4,729, belief in astrology by age, gender, region, religion |
| Google Trends (2023, cached) | State-level relative search interest for "horoscope" (scale 0–100) |
| US Census Bureau | State-level urbanization rate, college graduation %, median income, median age, total population |
Results
1. Who Believes: The Pew Picture
With 4,729 survey respondents, the national baseline is clear: 29% of Americans believe in astrology. But that aggregate conceals enormous variation by age, gender, and geography.
By age:
| Age Group | % Believe in Astrology |
|---|---|
| 18–29 | 44% |
| 30–49 | 32% |
| 50–64 | 22% |
| 65+ | 17% |
The age gradient is steeper than almost any other cultural belief measure. Belief is more than twice as common among young adults as among seniors. Whether this reflects a cohort effect (younger generations genuinely more drawn to astrology) or a life-stage effect (people grow out of it) cannot be determined from a single cross-sectional survey, but it's one of the most striking numbers in the dataset.
By gender:
| Gender | % Believe in Astrology |
|---|---|
| Women | 37% |
| Men | 20% |
A 17-percentage-point gap. This is consistent with decades of prior research on gender and metaphysical belief.
By region:
| Region | % Believe in Astrology |
|---|---|
| West | 33% |
| Northeast | 31% |
| South | 28% |
| Midwest | 25% |
The regional spread is modest (8 points coast to coast), but the direction is clear: coastal and urban-heavy regions show higher belief.
2. Where the Interest Is: State-Level Correlations
Using Google Trends search interest as a behavioral proxy, state-level data was correlated against Census demographic variables:
| Demographic Variable | Correlation (r) | p-value | Significant |
|---|---|---|---|
| Urban % | 0.719 | <0.0001 | Yes |
| Median income | 0.621 | <0.0001 | Yes |
| College graduation % | 0.620 | <0.0001 | Yes |
| Population (total) | 0.518 | 0.0001 | Yes |
| Median age | 0.214 | 0.135 | No |
Urbanization is the single strongest predictor — stronger than either income or education alone. A state's urban population percentage correlates with its astrological search interest at r = 0.719, which is a remarkably clean relationship for social science data.
A multiple regression model using urbanization, college graduation rate, and median income together explains 74.3% of the variance in state-level astrological search interest (R² = 0.743).
Methodological note: Urbanization, education, and income are themselves correlated with each other (multicollinearity). The model tells us these variables together predict astrological interest well; it cannot cleanly attribute the effect to any single one of them.
3. The Counterintuitive Finding
The most important finding in this project is what the correlations imply about astrology's cultural position.
Astrology interest is higher in more urban, more educated, higher-income states. This is the opposite of what a "folk superstition" framing would predict. If astrology were primarily a belief system of the rural, uneducated, or economically marginal — as it is sometimes stereotyped — we would expect negative correlations with education and income, and weak correlations with urbanization. We find the reverse.
This has a clear interpretation: in contemporary America, astrology functions as a culturally specific practice embedded in urban, educated, wellness-oriented social milieus — not as a traditional folk tradition persisting in less modernized communities. The primary driver appears to be social and cultural exposure. Cities concentrate the institutions (wellness culture, social media, progressive spiritual communities) through which astrology currently travels, and higher education and income track closely with urban residence.
This doesn't mean educated people are simply more credulous. It likely reflects a different kind of engagement: urban, educated users may be more drawn to astrology as a psychological and identity framework rather than as a literal predictive system. The Pew data supports this — belief rates are lower overall than search engagement would imply, suggesting that much of the Google Trends activity comes from curious non-believers.
4. Age as the Other Axis
The age gradient from Pew (44% at 18–29 vs. 17% at 65+) compounds the geographic pattern. Young people are disproportionately urban, and they're also disproportionately the ones who believe. Whether the future of astrological belief expands or contracts depends substantially on whether the current 18–29 cohort retains that belief as they age, or whether belief is a life-stage phenomenon that fades.
The data here cannot answer that — but it's the right question to carry forward.
Conclusion
Astrological interest in the United States is not randomly distributed. It concentrates in urban, younger, college-educated, higher-income populations, with women significantly overrepresented relative to men. The demographic variables together explain nearly three-quarters of the state-to-state variation in search interest.
The sociological conclusion is clear: astrology in 21st-century America is not a rural folk belief persisting at the margins of modernity. It is an urban cultural practice, most prevalent in the same populations that lead trends in wellness, spirituality, and identity-based consumer behavior. Understanding astrology's contemporary appeal requires understanding that social context — not just the content of astrological tradition itself.
Archived code, raw data outputs, and visualization (geographic_analysis.png) preserved in backup/.
Pew data source: Pew Research Center Religious Landscape Study (2017) and follow-up survey (2018).
Geographic Distribution of Astrological Consultations
Research Question
What are the geographic, demographic, and cultural patterns in interest and engagement with astrology? What factors predict higher or lower rates of astrological consultation?
Hypothesis
Interest in astrology varies systematically with geographic, demographic, and cultural factors, forming identifiable clusters that can be predicted by socioeconomic and cultural variables.
Background
Understanding who engages with astrology and where can inform both cultural studies and service provision. This research maps the landscape of astrological interest using digital trace data and demographic information.
Data Sources
- Google Trends: Search interest data for astrology-related terms
- Survey data: Pew Research Center, General Social Survey attitudes
- Census information: US Census Bureau demographic data
- Social media data: Reddit, Twitter engagement with astrology content
Mathematical Methods
- Spatial statistics: Geographic clustering analysis
- Clustering algorithms: K-means, DBSCAN for pattern identification
- Demographic modeling: Regression with demographic predictors
- Time-series analysis: Trends over time by region
Implementation Plan
Step 1: Data Collection
- Query Google Trends API for astrology-related searches by region
- Download relevant survey data from Pew/GSS
- Obtain census demographic data by region
- Collect Reddit/Twitter engagement metrics
Step 2: Data Preprocessing
- Normalize search data by population
- Merge datasets by geographic identifier
- Create derived demographic variables
- Handle missing data and outliers
Step 3: Spatial Analysis
- Map astrology interest geographically
- Calculate spatial autocorrelation (Moran's I)
- Identify hot spots and cold spots
- Test for clustering patterns
Step 4: Demographic Modeling
- Fit regression models with demographic predictors
- Test for interactions (age × education, etc.)
- Perform cluster analysis on combined features
- Validate clusters across different data sources
Step 5: Temporal Analysis
- Examine trends over time
- Test for seasonal patterns
- Identify events that spike interest
Expected Outputs
- Maps of astrology interest by region
- Demographic profiles of high-interest areas
- Clustering analysis results
- Predictive model for astrology engagement
Required Python Libraries
pytrends
pandas
numpy
scipy
scikit-learn
geopandas
folium
matplotlib
seaborn
Ethical Considerations
- Use only aggregate public data
- Avoid reinforcing stereotypes
- Present findings neutrally without judgment
Data Provenance
Geographic Coordinates
- Source: GeoNames / Astro-Databank.
- Link: https://www.geonames.org
- Reference: Gauquelin Sector Analysis.