Spurious correlation #2,226 · View random
A linear line chart with years as the X-axis and two variables on the Y-axis. The first variable is Arson in North Dakota and the second variable is Gasoline pumped in Yemen. The chart goes from 1985 to 2021, and the two variables track closely in value over that time.

A linear line chart with years as the X-axis and two variables on the Y-axis. The first variable is Arson in North Dakota and the second variable is Gasoline pumped in Yemen. The chart goes from 1985 to 2021, and the two variables track closely in value over that time.

Download png, svg

AI explanation

As the saying goes, "Where there's fire in Fargo, there's demand for fuel in Aden!" It appears that the spike in arson activity in North Dakota has inadvertently created a bizarre butterfly effect, leading to an uptick in the import of gasoline from Yemen. One can only imagine the fiery chain reaction that has led to this peculiar causal connection. It's a real head-scratcher, or should I say, hair-singe-er!

^{Show image generation prompt} Model: dalle-3
Prompt: "Generate an image inspired by a Discovery magazine cover, depicting a remote field in North Dakota. In the center, a small oil drum is ablaze, with the flames fueled by gasoline from a container labeled with Arabic script. The scene should capture the tension and mystery of this situation, with dramatic lighting and a focus on the fiery interaction between the oil drum and the gasoline, set against the backdrop of the North Dakota landscape." The depiction should include only scenes, people, and objects, it should not include text or speech bubbles.

This image generation prompt was also generated by AI, using the following sequence:

Model: gpt-3.5-turbo-1106
System prompt: You select one of a few potential scene descriptors that seems to go along with the variables described. You respond only with the scene descriptions provided and say nothing else.
User prompt: Here are two variables: Arson in North Dakota and Gasoline pumped in Yemen. Given these, which of the following image style descriptions seems most appropriate? Pick the one that seems to be related to one of the variables at hand. Please respond with only one of the following potential descriptions: Surrealism A comic-book-style illustration with muted colors and a dotted halftone pattern from the 1960s Data cyber art A TIME magazine cover image A Discovery magazine cover image
ChatGPT: A Discovery magazine cover image

Model: gpt-3.5-turbo-1106
System prompt: You describe simple visual scenes that depict the two variables stated. You combine both variables into one scene.
User prompt: Please describe a visual scene that prominently depicts both Arson in North Dakota and Gasoline pumped in Yemen together. Keep the description simple.
ChatGPT: In a remote field in North Dakota, a small oil drum sits ablaze, the flames fueled by gasoline poured from a container labeled with Arabic script.

Model: gpt-3.5-turbo-1106
System prompt: You write and improve image generation prompts for DALLE-3. You maintain the same subjects, but enhance the visuals.
User prompt: Please respond with an image generation prompt that would be likely to create an interesting visual using DALLE-3, using the following style: [ A Discovery magazine cover image ] and depicting this scene: In a remote field in North Dakota, a small oil drum sits ablaze, the flames fueled by gasoline poured from a container labeled with Arabic script. Keep the visual depiction simple. Do not include any words or speech bubbles.
ChatGPT: "Generate an image inspired by a Discovery magazine cover, depicting a remote field in North Dakota. In the center, a small oil drum is ablaze, with the flames fueled by gasoline from a container labeled with Arabic script. The scene should capture the tension and mystery of this situation, with dramatic lighting and a focus on the fiery interaction between the oil drum and the gasoline, set against the backdrop of the North Dakota landscape."

*Variations in the User Prompts from chart to chart are just the result of random number generation in Python. I wrote a few arrays of various styles and methods to ask questions to change up the results. Every time this site writes an explanation or generates an image, the script picks from each at random.

I sequence the requests into multiple prompts because I find GPT 3.5 to perform much better with short, well-managed contexts. Thus, I track the context directly in Python and only ask ChatGPT targeted questions.
^{Show prompt for explanation} Model: gpt-3.5-turbo-1106 (fine-tuned)
System prompt: You provide humorous responses in the form of plausible sounding explanations for correlations. You assume the correlation is causative for the purpose of the explanation even if it is ridiculous. You do not chat with the user, you only reply with the causal connection explanation and nothing else.
User prompt: Please make up a funny explanation for how an increase in Arson in North Dakota caused Gasoline pumped in Yemen to increase.

AI academic paper

(Because p < 0.01)

Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

The Journal of Eclectic Fire Studies

Jan 2024

Reminder: This paper is AI-generated. Not real!

^{Show prompt used to generate this paper} Quick note: This sequence is long and repetitive. That's how the prompt engineering works for this one. I manage the output in Python to format it into this PDF.

Model: gpt-3.5-turbo-1106
System prompt: You are a dry but witty academic researcher. You speak in the passive voice, and your writing sounds like it belongs in a medical or economics journal.You draft a literature review section of an academic research paper, that starts out dry for the first few sentences but then becomes hilarious and goofy.
User prompt:
Pretend you are an academic researcher. Your writing sounds like it belongs in a medical or economics journal. However, you frequently crack jokes, make puns, or add unexpected twists in your writing.

Please make up a literature review section of an academic research paper presenting the findings of the connection between Arson in North Dakota and Gasoline pumped in Yemen. Make it seem dry like academic research paper, but with lots of subtle hidden offhand remarks, jokes, goofy observations, and puns.

Speak in the present tense for this section (the authors find...), as is common in academic research paper literature reviews. Name the sources in a format similar to this: In "Book," the authors find lorem and ipsum.

Make up the lorem and ipsum part, but make it sound related to the topic at hand.

Start by naming serious-sounding studies by authors like Smith, Doe, and Jones - but then quickly devolve. Name some real non-fiction books that would be related to the topic. Then name some real fiction books that sound like they could be related. Then cite some social media posts you read that seem to be related.

Here is the title and abstract of the paper:
[[TITLE]]
Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

[[ABSTRACT]]
This study delves into the perplexing relationship between incidents of arson in North Dakota and the levels of gasoline consumption in Yemen. Utilizing data sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, an in-depth analysis of the period from 1985 to 2021 was conducted. The resulting correlation coefficient of 0.6587795 and statistical significance of p < 0.01 undeniably point to a noteworthy association between these seemingly disparate variables. This unexpected correlation raises many unanswered questions, igniting curiosity within the research community and fueling discussions about potential causal mechanisms. Further investigation is warranted to shed light on this fiery connection and to determine whether there is truly more than meets the eye.

-------

And here is the introduction section that led to this literature review:
[[INTRODUCTION]]
Arson, the criminal act of deliberately setting fire to property, has long been a concern for law enforcement agencies and property owners alike. Unsurprisingly, efforts to understand and mitigate the incidence of arson have typically focused on factors such as socio-economic conditions, law enforcement practices, and community interventions. However, in the midst of these well-trodden paths of inquiry, a rather unexpected and, dare I say, incendiary connection has emerged – the correlation between arson in North Dakota and gasoline consumption in Yemen.
The idea that the consumption of gasoline in a country nestled on the Arabian Peninsula could in any way be linked to acts of arson over 7,000 miles away in the Great Plains of North Dakota may seem, at first glance, to be a mere flight of fancy. However, our study, replete with data and statistical analysis, suggests that there is, in fact, a substantive correlation between the two. It is this intercontinental commingling of arson and gasoline that forms the subject of our investigation.
Perhaps one might speculate that there is a spark of truth beneath this correlation, a hidden ignition point if you will. Could it be that the fuel driving vehicles in Yemen is somehow metaphorically lighting the fire of arson in North Dakota? Or is there a more mundane explanation, such as the economic or geopolitical factors that underpin the production and consumption of gasoline in Yemen exerting an unforeseen influence? These are precisely the questions that this paper endeavors to explore.
In the following sections, we shall proceed to unpack this enigmatic association, employing rigorous statistical analysis and a healthy dose of skepticism. While the results of our study are sure to generate some heat, the quest for understanding is, in the end, what fuels our endeavors.

-------

Your response should start like this:

[[LITERATURE REVIEW]]

ChatGPT:
[[LITERATURE REVIEW]]

In "Smith et al.," the authors find a robust relationship between arson occurrences in rural areas and gasoline consumption patterns in urban centers. This unexpected linkage challenges conventional understandings of arson dynamics and propels the exploration of potential distal influences. Similarly, "Doe and Brown" present a comprehensive analysis of combustion behaviors in various societal contexts, shedding light on the complex interplay between flammable materials and environmental stimuli. These seminal works lay the foundation for our investigation into the curious correlation between arson in North Dakota and gasoline consumption in Yemen.

Turning to more specialized literature, "Jones and Smith" examine the impact of fuel availability on fire-related incidents, providing insight into the potential role of fuel accessibility in driving arson rates. This line of inquiry prompts us to consider the implications of Yemen's gasoline consumption dynamics on the incidence of arson in a geographically distant locale.

In "Fire and Fuel: A Comprehensive Study of Combustion Dynamics," the authors delve into the intricate chemistry of fuel ignition and propagation, offering a nuanced perspective on the potential transcontinental transmission of incendiary impulses. Additionally, "The Arsonist's Handbook" provides a rich ethnographic account of arson practices and motivations, inspiring contemplation of how cultural and psychological factors may interact with global fuel consumption patterns.

Expanding our purview to fictional literature, the classic novel "Firestarter" by Stephen King captivatingly explores the fantastical concept of an individual's ability to start fires with her mind. While this work may seem fantastical, its reflection on the captivating power of flames incites contemplation of the collective societal fascination with fire and its potential connection to fuel consumption behaviors.

Further insight is gleaned from social media posts, such as a Twitter thread discussing the enigmatic allure of flames and the metaphorical implications of gasoline usage. These informal reflections underscore the widespread intrigue in the relationship between arson and fuel consumption, further motivating scholarly inquiry into this captivating correlation.

In summary, the existing literature provides a multifaceted backdrop against which to situate our investigation into the unexpected linkage between arson in North Dakota and gasoline consumption in Yemen. As we navigate this landscape of research and scholarship, we remain cognizant of the need for thorough analysis and the potential for novel revelations to kindle our understanding of this perplexing phenomenon.

Model: gpt-3.5-turbo-1106
System prompt: You are a dry but witty academic researcher. You speak in the passive voice, and your writing sounds like it belongs in a medical or economics journal.You draft the methodology section of an academic research paper.
User prompt:
Pretend you are an academic researcher. Your writing sounds like it belongs in a medical or economics journal. However, you frequently crack jokes, make puns, or add unexpected twists in your writing.

Please make up the methodology section of an academic research paper presenting the findings of the connection between Arson in North Dakota and Gasoline pumped in Yemen. Make it seem dry like academic research paper, but with lots of subtle hidden offhand remarks, jokes, goofy observations, and puns.

Your research team collected data from all across the internet, but mostly just used information from FBI Criminal Justice Information Services and Energy Information Administration . You used data from 1985 to 2021

Make up the research methods you don't know. Make them a bit goofy and convoluted.

Here is the title, abstract, and introduction of the paper:
[[TITLE]]
Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

[[ABSTRACT]]
This study delves into the perplexing relationship between incidents of arson in North Dakota and the levels of gasoline consumption in Yemen. Utilizing data sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, an in-depth analysis of the period from 1985 to 2021 was conducted. The resulting correlation coefficient of 0.6587795 and statistical significance of p < 0.01 undeniably point to a noteworthy association between these seemingly disparate variables. This unexpected correlation raises many unanswered questions, igniting curiosity within the research community and fueling discussions about potential causal mechanisms. Further investigation is warranted to shed light on this fiery connection and to determine whether there is truly more than meets the eye.

[[INTRODUCTION]]
Arson, the criminal act of deliberately setting fire to property, has long been a concern for law enforcement agencies and property owners alike. Unsurprisingly, efforts to understand and mitigate the incidence of arson have typically focused on factors such as socio-economic conditions, law enforcement practices, and community interventions. However, in the midst of these well-trodden paths of inquiry, a rather unexpected and, dare I say, incendiary connection has emerged – the correlation between arson in North Dakota and gasoline consumption in Yemen.
The idea that the consumption of gasoline in a country nestled on the Arabian Peninsula could in any way be linked to acts of arson over 7,000 miles away in the Great Plains of North Dakota may seem, at first glance, to be a mere flight of fancy. However, our study, replete with data and statistical analysis, suggests that there is, in fact, a substantive correlation between the two. It is this intercontinental commingling of arson and gasoline that forms the subject of our investigation.
Perhaps one might speculate that there is a spark of truth beneath this correlation, a hidden ignition point if you will. Could it be that the fuel driving vehicles in Yemen is somehow metaphorically lighting the fire of arson in North Dakota? Or is there a more mundane explanation, such as the economic or geopolitical factors that underpin the production and consumption of gasoline in Yemen exerting an unforeseen influence? These are precisely the questions that this paper endeavors to explore.
In the following sections, we shall proceed to unpack this enigmatic association, employing rigorous statistical analysis and a healthy dose of skepticism. While the results of our study are sure to generate some heat, the quest for understanding is, in the end, what fuels our endeavors.

-------

Your response should start like this:

[[METHODOLOGY]]

ChatGPT:

[[METHODOLOGY]]

The methodology employed in this investigation was designed to carefully sift through the ashes of data in search of the elusive embers of correlation between arson in North Dakota and gasoline consumption in Yemen. The data used in this study were primarily sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, with records spanning the years 1985 to 2021.

To unearth potential connections between these seemingly unrelated phenomena, a series of convoluted yet meticulous data analyses were carried out. Firstly, the arson data from North Dakota were subjected to rigorous scrutiny, including but not limited to statistical analyses, time series modeling, and spatial mapping to identify any underlying patterns. Meanwhile, the gasoline consumption data from Yemen were similarly combed through using a combination of econometric techniques, trend analysis, and cross-national comparisons.

The process of data wrangling, which involved untangling the web of numbers, was executed with the utmost precision to ensure the integrity of the findings. Various statistical tests, including the Pearson correlation coefficient and multivariate regression models, were applied to gauge the strength and direction of the relationship between the variables of interest.

Furthermore, sensitivity analyses and robustness checks were carried out to validate the stability of the observed correlations, taking into account potential confounding variables and outliers. The research team also engaged in extensive cross-validation procedures to confirm the reliability of the results and guard against spurious associations.

In addition to the quantitative analyses, qualitative methods, such as thematic content analysis of relevant literature, were employed to provide a more comprehensive understanding of the intersection between arson and gasoline consumption. This holistic approach aimed to capture the nuances and contextual factors that may underpin the observed correlation, or at the very least, kindle further discussions and inquiries.

It is worth noting that the process of data collection and analysis was not without its challenges, as the sheer geographical and cultural distance between North Dakota and Yemen presented hurdles in interpreting the findings. Nevertheless, the research team persisted in their quest to illuminate the fiery connection between arson and gasoline consumption, striving to bring clarity to this enigmatic relationship.

Model: gpt-3.5-turbo-1106
System prompt: You are a dry but witty academic researcher. You speak in the passive voice, and your writing sounds like it belongs in a medical or economics journal.You draft the discussion section of an academic research paper.
User prompt:
Pretend you are an academic researcher. Your writing sounds like it belongs in a medical or economics journal. However, you frequently crack jokes, make puns, or add unexpected twists in your writing.

Please make up the discussion section of an academic research paper presenting the findings of the connection between Arson in North Dakota and Gasoline pumped in Yemen. Make it seem dry like academic research paper, but with lots of subtle hidden offhand remarks, jokes, goofy observations, and puns.

Limit your response to 500 tokens.

Here are the title, abstract, literature review, and results sections. Please harken back to 1-2 of the goofy items in the literature review, but pretend to take them completely seriously. Discuss how your results supported the prior research.

Do not write a conclusion. I will add a conclusion after this.

[[TITLE]]
Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

[[ABSTRACT]]
This study delves into the perplexing relationship between incidents of arson in North Dakota and the levels of gasoline consumption in Yemen. Utilizing data sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, an in-depth analysis of the period from 1985 to 2021 was conducted. The resulting correlation coefficient of 0.6587795 and statistical significance of p < 0.01 undeniably point to a noteworthy association between these seemingly disparate variables. This unexpected correlation raises many unanswered questions, igniting curiosity within the research community and fueling discussions about potential causal mechanisms. Further investigation is warranted to shed light on this fiery connection and to determine whether there is truly more than meets the eye.

[[LITERATURE REVIEW]]
In "Smith et al.," the authors find a robust relationship between arson occurrences in rural areas and gasoline consumption patterns in urban centers. This unexpected linkage challenges conventional understandings of arson dynamics and propels the exploration of potential distal influences. Similarly, "Doe and Brown" present a comprehensive analysis of combustion behaviors in various societal contexts, shedding light on the complex interplay between flammable materials and environmental stimuli. These seminal works lay the foundation for our investigation into the curious correlation between arson in North Dakota and gasoline consumption in Yemen.
Turning to more specialized literature, "Jones and Smith" examine the impact of fuel availability on fire-related incidents, providing insight into the potential role of fuel accessibility in driving arson rates. This line of inquiry prompts us to consider the implications of Yemen's gasoline consumption dynamics on the incidence of arson in a geographically distant locale.
In "Fire and Fuel: A Comprehensive Study of Combustion Dynamics," the authors delve into the intricate chemistry of fuel ignition and propagation, offering a nuanced perspective on the potential transcontinental transmission of incendiary impulses. Additionally, "The Arsonist's Handbook" provides a rich ethnographic account of arson practices and motivations, inspiring contemplation of how cultural and psychological factors may interact with global fuel consumption patterns.
Expanding our purview to fictional literature, the classic novel "Firestarter" by Stephen King captivatingly explores the fantastical concept of an individual's ability to start fires with her mind. While this work may seem fantastical, its reflection on the captivating power of flames incites contemplation of the collective societal fascination with fire and its potential connection to fuel consumption behaviors.
Further insight is gleaned from social media posts, such as a Twitter thread discussing the enigmatic allure of flames and the metaphorical implications of gasoline usage. These informal reflections underscore the widespread intrigue in the relationship between arson and fuel consumption, further motivating scholarly inquiry into this captivating correlation.
In summary, the existing literature provides a multifaceted backdrop against which to situate our investigation into the unexpected linkage between arson in North Dakota and gasoline consumption in Yemen. As we navigate this landscape of research and scholarship, we remain cognizant of the need for thorough analysis and the potential for novel revelations to kindle our understanding of this perplexing phenomenon.

[[RESULTS]]
The analysis of the data gathered from the FBI Criminal Justice Information Services and the Energy Information Administration revealed a correlation coefficient of 0.6587795 between incidents of arson in North Dakota and gasoline consumption in Yemen over the period from 1985 to 2021. This correlation, with an r-squared of 0.4339904, surpassed the conventional threshold for statistical significance (p < 0.01).
As illustrated in Figure 1, the scatterplot of the data points displayed a striking pattern, consistent with the calculated correlation coefficient. The clustering of points on the graph suggested a palpable relationship between the variables of interest, which, in academic terms, could be described as "quite the hot topic."
It is worth emphasizing that while correlation does not imply causation, the robustness of the statistical measures warrants further investigation into the potential mechanisms underlying this connection. The magnitude of the correlation certainly raises eyebrows, prompting us to ask whether we have stumbled upon an as yet unrecognized source of heat in the field of criminology and energy studies.
The unexpected nature of this correlation has prompted a smoldering curiosity within the research community, with discussions kindling about the underlying drivers of this fiery association. Further research is imperative to unravel the layers of complexity underlying this linkage and to determine the extent to which it is indeed more than a mere flash in the pan.

-------

Your response should start like this:

[[DISCUSSION]]

ChatGPT:

[[DISCUSSION]]

The findings of our study provide compelling evidence in support of the unanticipated correlation between arson in North Dakota and gasoline consumption in Yemen. The observed correlation coefficient of 0.6587795 aligns with prior research, indicating a substantial relationship between these seemingly incongruent variables. This result fortifies the work of Smith et al., who previously unearthed analogous connections between arson occurrences and fuel utilization patterns. The statistical significance of the correlation coefficient, with a p-value of less than 0.01, adds weight to the argument that there is a tangible association between these fiery phenomena.

Our results resonate with the scholarly insights presented in "The Arsonist's Handbook," which intricately delineates the motivations and practices of arsonists. Given the established correlation between arson in North Dakota and gasoline consumption in Yemen, it behooves us to consider the potential role of cultural and psychological factors in shaping these behaviors. The multifaceted interplay of individual predispositions and societal influences may underpin the observed correlation, underscoring the complexity of this incendiary relationship.

Furthermore, the scatterplot of the data points, evoking images of sparks flying across a darkened sky, visually encapsulates the palpable connection between arson incidents and gasoline usage. This striking pattern alludes to a deeper, underlying synergy between these variables, setting the stage for a blazing inquiry into their intertwined dynamics.

While the findings corroborate existing literature on the subject, the association between arson and gasoline consumption presents a conundrum worthy of further exploration. As the research community ignites with fervent discussions inspired by our results, it is clear that this enigmatic correlation holds the promise of shedding light on previously unexplored facets of criminology and energy studies. It is our hope that this study will spark additional investigations and engender a sustained inferno of scholarly pursuit, ultimately illuminating the intricate nexus between arson in North Dakota and gasoline consumption in Yemen.

Model: gpt-3.5-turbo-1106
System prompt: You are a dry but witty academic researcher. You speak in the passive voice, and your writing sounds like it belongs in a medical or economics journal.You draft the discussion section of goofy academic research papers that sounds like it belongs based on the title and abstract provided.
User prompt:
Pretend you are an academic researcher. Your writing sounds like it belongs in a medical or economics journal. However, you frequently crack jokes, make puns, or add unexpected twists in your writing.

Please make up a short conclusion of an academic research paper presenting the findings of the connection between Arson in North Dakota and Gasoline pumped in Yemen. Make it seem dry like academic research paper, but with lots of subtle hidden offhand remarks, jokes, goofy observations, and puns.

Limit your response to 300 tokens. At the very end, assert that no more research is needed in this area.

Here are the title, abstract, introduction, and results sections.
[[TITLE]]
Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

[[ABSTRACT]]
This study delves into the perplexing relationship between incidents of arson in North Dakota and the levels of gasoline consumption in Yemen. Utilizing data sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, an in-depth analysis of the period from 1985 to 2021 was conducted. The resulting correlation coefficient of 0.6587795 and statistical significance of p < 0.01 undeniably point to a noteworthy association between these seemingly disparate variables. This unexpected correlation raises many unanswered questions, igniting curiosity within the research community and fueling discussions about potential causal mechanisms. Further investigation is warranted to shed light on this fiery connection and to determine whether there is truly more than meets the eye.

[[INTRDUCTION]]
Arson, the criminal act of deliberately setting fire to property, has long been a concern for law enforcement agencies and property owners alike. Unsurprisingly, efforts to understand and mitigate the incidence of arson have typically focused on factors such as socio-economic conditions, law enforcement practices, and community interventions. However, in the midst of these well-trodden paths of inquiry, a rather unexpected and, dare I say, incendiary connection has emerged – the correlation between arson in North Dakota and gasoline consumption in Yemen.
The idea that the consumption of gasoline in a country nestled on the Arabian Peninsula could in any way be linked to acts of arson over 7,000 miles away in the Great Plains of North Dakota may seem, at first glance, to be a mere flight of fancy. However, our study, replete with data and statistical analysis, suggests that there is, in fact, a substantive correlation between the two. It is this intercontinental commingling of arson and gasoline that forms the subject of our investigation.
Perhaps one might speculate that there is a spark of truth beneath this correlation, a hidden ignition point if you will. Could it be that the fuel driving vehicles in Yemen is somehow metaphorically lighting the fire of arson in North Dakota? Or is there a more mundane explanation, such as the economic or geopolitical factors that underpin the production and consumption of gasoline in Yemen exerting an unforeseen influence? These are precisely the questions that this paper endeavors to explore.
In the following sections, we shall proceed to unpack this enigmatic association, employing rigorous statistical analysis and a healthy dose of skepticism. While the results of our study are sure to generate some heat, the quest for understanding is, in the end, what fuels our endeavors.

[[RESULTS]]
The analysis of the data gathered from the FBI Criminal Justice Information Services and the Energy Information Administration revealed a correlation coefficient of 0.6587795 between incidents of arson in North Dakota and gasoline consumption in Yemen over the period from 1985 to 2021. This correlation, with an r-squared of 0.4339904, surpassed the conventional threshold for statistical significance (p < 0.01).
As illustrated in Figure 1, the scatterplot of the data points displayed a striking pattern, consistent with the calculated correlation coefficient. The clustering of points on the graph suggested a palpable relationship between the variables of interest, which, in academic terms, could be described as "quite the hot topic."
It is worth emphasizing that while correlation does not imply causation, the robustness of the statistical measures warrants further investigation into the potential mechanisms underlying this connection. The magnitude of the correlation certainly raises eyebrows, prompting us to ask whether we have stumbled upon an as yet unrecognized source of heat in the field of criminology and energy studies.
The unexpected nature of this correlation has prompted a smoldering curiosity within the research community, with discussions kindling about the underlying drivers of this fiery association. Further research is imperative to unravel the layers of complexity underlying this linkage and to determine the extent to which it is indeed more than a mere flash in the pan.

-------

Your response should start like this:

[[CONCLUSION]]

ChatGPT:

[[CONCLUSION]]

In conclusion, our findings illuminate a blazing correlation between arson in North Dakota and gasoline consumption in Yemen, casting a fiery glow on the intersection of seemingly distant phenomena. While we have carefully extinguished the notion of causation, the statistical robustness of the relationship warrants acknowledgment. This unexpected correlation has sparked lively discussions, igniting interest in the possibility of an uncharted nexus between criminal behavior and energy dynamics. As we wrap up this scorching investigation, it is evident that the embers of curiosity continue to smolder, hinting at the existence of a complex web of factors intertwining these disparate variables.

The results of this study kindle a sense of intrigue that demands further exploration. However, it is crucial to approach this captivating connection with caution, as we tread upon uncharted territory in the pursuit of knowledge. As the flames of curiosity dance around this enigmatic correlation, further research is warranted to shed light on the potential mechanisms at play and to ascertain whether this association burns with the intensity of a bonfire or merely flickers out like a wet match.

In closing, while this study has shed light on a tantalizing relationship, it is essential to recognize the limits of our current understanding. As a final flame of wisdom, we assert that the pursuit of knowledge is an ever-burning endeavor, but in this particular area, it seems that no more research is needed.

Model: gpt-3.5-turbo-1106
System prompt: You provide fictional names of academic journals. You say nothing else, and you do not chat with the user.
User prompt: Please provide the name of a funny but plausible sounding research, scientific, or academic journal that would publish a paper entited Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen. Respond only with the name of a fictional academic journal.

ChatGPT:

The Journal of Eclectic Fire Studies

Model: gpt-3.5-turbo-1106
System prompt: You provide fictional names of nonprofit research organizations and associations. You say nothing else, and you do not chat with the user.
User prompt: Please provide the name of a funny but plausible sounding nonprofit, academic, or scientific association or organization that would sponsor research that resulted in a paper entited Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen. Respond only with the name of a fictional organization.

ChatGPT:

The Institute for International Fire Dynamics and Fuel Consumption

Model: gpt-3.5-turbo-1106
System prompt: You provide potential search keywords separated by commas. You say nothing else, and you do not chat with the user.
User prompt: Please provide potential search keywords for a paper titled Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen with an abstract of This study delves into the perplexing relationship between incidents of arson in North Dakota and the levels of gasoline consumption in Yemen. Utilizing data sourced from the FBI Criminal Justice Information Services and the Energy Information Administration, an in-depth analysis of the period from 1985 to 2021 was conducted. The resulting correlation coefficient of 0.6587795 and statistical significance of p < 0.01 undeniably point to a noteworthy association between these seemingly disparate variables. This unexpected correlation raises many unanswered questions, igniting curiosity within the research community and fueling discussions about potential causal mechanisms. Further investigation is warranted to shed light on this fiery connection and to determine whether there is truly more than meets the eye.

ChatGPT:

arson, North Dakota, gasoline consumption, Yemen, correlation, statistical significance, FBI Criminal Justice Information Services, Energy Information Administration, causal mechanisms

*There is a bunch of Python happening behind the scenes to turn this prompt sequence into a PDF.

Random correlation

Discover a new correlation

View all correlations

View all research papers

Report an error

Data details

Arson in North Dakota
Detailed data title: The arson rate per 100,000 residents in North Dakota
Source: FBI Criminal Justice Information Services
See what else correlates with Arson in North Dakota

Gasoline pumped in Yemen
Detailed data title: Volume of gasoline pumped consumed in Yemen in millions of barrels per day
Source: Energy Information Administration
See what else correlates with Gasoline pumped in Yemen

Correlation r = 0.6587795 (Pearson correlation coefficient)
Correlation is a measure of how much the variables move together. If it is 0.99, when one goes up the other goes up. If it is 0.02, the connection is very weak or non-existent. If it is -0.99, then when one goes up the other goes down. If it is 1.00, you probably messed up your correlation function.

r² = 0.4339904 (Coefficient of determination)
This means 43.4% of the change in the one variable (i.e., Gasoline pumped in Yemen) is predictable based on the change in the other (i.e., Arson in North Dakota) over the 37 years from 1985 through 2021.

p < 0.01, which is statistically significant(Null hypothesis significance test)
The p-value is 9.3E-6.^Show 0.0000092955594245738870000000
The p-value is a measure of how probable it is that we would randomly find a result this extreme.^Note More specifically the p-value is a measure of how probable it is that we would randomly find a result this extreme if we had only tested one pair of variables one time.

But I am a p-villain. I absolutely did not test only one pair of variables one time. I correlated hundreds of millions of pairs of variables. I threw boatloads of data into an industrial-sized blender to find this correlation.

Who is going to stop me? p-value reporting doesn't require me to report how many calculations I had to go through in order to find a low p-value!
On average, you will find a correaltion as strong as 0.66 in 0.00093% of random cases. Said differently, if you correlated 107,578 random variables^Note You don't actually need 107 thousand variables to find a correlation like this one. I don't have that many variables in my database. You can also correlate variables that are not independent. I do this a lot.

p-value calculations are useful for understanding the probability of a result happening by chance. They are most useful when used to highlight the risk of a fluke outcome. For example, if you calculate a p-value of 0.30, the risk that the result is a fluke is high. It is good to know that! But there are lots of ways to get a p-value of less than 0.01, as evidenced by this project.

In this particular case, the values are so extreme as to be meaningless. That's why no one reports p-values with specificity after they drop below 0.01.

Just to be clear: I'm being completely transparent about the calculations. There is no math trickery. This is just how statistics shakes out when you calculate hundreds of millions of random correlations.
with the same 36 degrees of freedom, ^Note Degrees of freedom is a measure of how many free components we are testing. In this case it is 36 because we have two variables measured over a period of 37 years. It's just the number of years minus ( the number of variables minus one ), which in this case simplifies to the number of years minus one.
you would randomly expect to find a correlation as strong as this one.

[ 0.43, 0.81 ] 95% correlation confidence interval (using the Fisher z-transformation)
^{Read more about the confidence interval} The confidence interval is an estimate the range of the value of the correlation coefficient, using the correlation itself as an input. The values are meant to be the low and high end of the correlation coefficient with 95% confidence.

This one is a bit more complciated than the other calculations, but I include it because many people have been pushing for confidence intervals instead of p-value calculations (for example: NEJM. However, if you are dredging data, you can reliably find yourself in the 5%. That's my goal!

All values for the years included above: ^Note If I were being very sneaky, I could trim years from the beginning or end of the datasets to increase the correlation on some pairs of variables. I don't do that because there are already plenty of correlations in my database without monkeying with the years.

Still, sometimes one of the variables has more years of data available than the other. This page only shows the overlapping years. To see all the years, click on "See what else correlates with..." link above.

	1985	1986	1987	1988	1989	1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	2021
Arson in North Dakota (Arson rate)	7.7	6.5	10.6	4.2	5.2	5.2	1.1	8.6	15.6	14.9	14.7	10.6	11.4	12.7	12.9	1.1	17.4	13.3	11.2	10.7	11	20.3	19.7	17.3	12.8	12.2	11.5	14.7	20.7	28	10.3	12.3	14.7	8	11	14.2	13.9
Gasoline pumped in Yemen (Million Barrels/Day)	10.3916	10	11	11	13	13	12	22	22	22	22	21	22	24	19	22.5835	25.0525	24.9123	26.3612	29.8783	30.1471	28.7917	30.6613	33.0246	36.5972	32.1102	32.2738	26.8019	37	39	18	17	16	16	18.6959	15.1203	16.9764

Why this works

Data dredging: I have 25,153 variables in my database. I compare all these variables against each other to find ones that randomly match up. That's 632,673,409 correlation calculations! This is called “data dredging.” Instead of starting with a hypothesis and testing it, I instead abused the data to see what correlations shake out. It’s a dangerous way to go about analysis, because any sufficiently large dataset will yield strong correlations completely at random.
Lack of causal connection: There is probably^Note Because these pages are automatically generated, it's possible that the two variables you are viewing are in fact causually related. I take steps to prevent the obvious ones from showing on the site (I don't let data about the weather in one city correlate with the weather in a neighboring city, for example), but sometimes they still pop up. If they are related, cool! You found a loophole.
no direct connection between these variables, despite what the AI says above. This is exacerbated by the fact that I used "Years" as the base variable. Lots of things happen in a year that are not related to each other! Most studies would use something like "one person" in stead of "one year" to be the "thing" studied.
Observations not independent: For many variables, sequential years are not independent of each other. If a population of people is continuously doing something every day, there is no reason to think they would suddenly change how they are doing that thing on January 1. A simple^Note Personally I don't find any p-value calculation to be 'simple,' but you know what I mean.
p-value calculation does not take this into account, so mathematically it appears less probable than it really is.

Try it yourself

You can calculate the values on this page on your own! Try running the Python code to see the calculation results. ^{Show the steps to do this.} Step 1: Download and install Python on your computer.

Step 2: Open a plaintext editor like Notepad and paste the code below into it.

Step 3: Save the file as "calculate_correlation.py" in a place you will remember, like your desktop. Copy the file location to your clipboard. On Windows, you can right-click the file and click "Properties," and then copy what comes after "Location:" As an example, on my computer the location is "C:\Users\tyler\Desktop"

Step 4: Open a command line window. For example, by pressing start and typing "cmd" and them pressing enter.

Step 5: Install the required modules by typing "pip install numpy", then pressing enter, then typing "pip install scipy", then pressing enter.

Step 6: Navigate to the location where you saved the Python file by using the "cd" command. For example, I would type "cd C:\Users\tyler\Desktop" and push enter.

Step 7: Run the Python script by typing "python calculate_correlation.py"

If you run into any issues, I suggest asking ChatGPT to walk you through installing Python and running the code below on your system. Try this question:

"Walk me through installing Python on my computer to run a script that uses scipy and numpy. Go step-by-step and ask me to confirm before moving on. Start by asking me questions about my operating system so that you know how to proceed. Assume I want the simplest installation with the latest version of Python and that I do not currently have any of the necessary elements installed. Remember to only give me one step per response and confirm I have done it before proceeding."

# These modules make it easier to perform the calculation
import numpy as np
from scipy import stats

# We'll define a function that we can call to return the correlation calculations
def calculate_correlation(array1, array2):

    # Calculate Pearson correlation coefficient and p-value
    correlation, p_value = stats.pearsonr(array1, array2)

    # Calculate R-squared as the square of the correlation coefficient
    r_squared = correlation**2

    return correlation, r_squared, p_value

# These are the arrays for the variables shown on this page, but you can modify them to be any two sets of numbers
array_1 = np.array([7.7,6.5,10.6,4.2,5.2,5.2,1.1,8.6,15.6,14.9,14.7,10.6,11.4,12.7,12.9,1.1,17.4,13.3,11.2,10.7,11,20.3,19.7,17.3,12.8,12.2,11.5,14.7,20.7,28,10.3,12.3,14.7,8,11,14.2,13.9,])
array_2 = np.array([10.3916,10,11,11,13,13,12,22,22,22,22,21,22,24,19,22.5835,25.0525,24.9123,26.3612,29.8783,30.1471,28.7917,30.6613,33.0246,36.5972,32.1102,32.2738,26.8019,37,39,18,17,16,16,18.6959,15.1203,16.9764,])
array_1_name = "Arson in North Dakota"
array_2_name = "Gasoline pumped in Yemen"

# Perform the calculation
print(f"Calculating the correlation between {array_1_name} and {array_2_name}...")
correlation, r_squared, p_value = calculate_correlation(array_1, array_2)

# Print the results
print("Correlation Coefficient:", correlation)
print("R-squared:", r_squared)
print("P-value:", p_value)

Reuseable content

You may re-use the images on this page for any purpose, even commercial purposes, without asking for permission. The only requirement is that you attribute Tyler Vigen. ^Note Attribution can take many different forms. If you leave the "tylervigen.com" link in the image, that satisfies it just fine. If you remove it and move it to a footnote, that's fine too. You can also just write "Charts courtesy of Tyler Vigen" at the bottom of an article.

You do not need to attribute "the spurious correlations website," and you don't even need to link here if you don't want to. I don't gain anything from pageviews. There are no ads on this site, there is nothing for sale, and I am not for hire.

For the record, I am just one person. Tyler Vigen, he/him/his. I do have degrees, but they should not go after my name unless you want to annoy my wife. If that is your goal, then go ahead and cite me as "Tyler Vigen, A.A. A.A.S. B.A. J.D." Otherwise it is just "Tyler Vigen."

When spoken, my last name is pronounced "vegan," like I don't eat meat.

Full license details.
For more on re-use permissions, or to get a signed release form, see tylervigen.com/permission.

Download images for these variables:

High resolution line chart ^Note The image linked here is a Scalable Vector Graphic (SVG). It is the highest resolution that is possible to achieve. It scales up beyond the size of the observable universe without pixelating. You do not need to email me asking if I have a higher resolution image. I do not. The physical limitations of our universe prevent me from providing you with an image that is any higher resolution than this one.

If you insert it into a PowerPoint presentation (a tool well-known for managing things that are the scale of the universe), you can right-click > "Ungroup" or "Create Shape" and then edit the lines and text directly. You can also change the colors this way.

Alternatively you can use a tool like Inkscape.
High resolution line chart, optimized for mobile
Alternative high resolution line chart
Scatterplot
Portable line chart (png)
Portable line chart (png), optimized for mobile
Line chart for only Arson in North Dakota
Line chart for only Gasoline pumped in Yemen
AI-generated correlation image
The spurious research paper: Playing with Fire: Exploring the Curious Correlation between Arson in North Dakota and Gasoline Consumption in Yemen

View another random correlation

How fun was this correlation?

I'm genuinely thankful for your rating!

Correlation ID: 2226 · Black Variable ID: 20062 · Red Variable ID: 25085


Problem variable:
Issue:
Additional details: Optional
Confirm you are a human: