# The Fallacy of the Null-Hypothesis Significance Test Quiz Questions and Answers

What is your absolute favorite anecdote from The Fallacy of the Null-Hypothesis Significance Test and why?

• The graduate student’s dissertation because it perfectly illustrates the real-world implications of flawed statistical thinking.
• The inveterate gambler because it clarifies the concept of “degree of belief” in a relatable way.
• Honestly, I skimmed the anecdotes and went straight for the statistical arguments.

What is your current level of expertise in probabilistic inference?

• Beginner: I’m just starting to understand the basics.
• Intermediate: I’m familiar with the concepts, but still learning how to apply them.
• Expert: I use probabilistic inference regularly in my work.

How often do you question the validity of widely accepted methods in your field?

• Frequently: I’m always looking for flaws and alternative approaches.
• Occasionally: I question things when I encounter inconsistencies or limitations.
• Rarely: I trust that established methods have been thoroughly vetted.

How well do you think most researchers understand the concepts of “probability” and “hypothesis likelihood”?

• Very well: These concepts are fundamental to scientific reasoning.
• Somewhat well: There’s always room for improvement in understanding these complex ideas.
• Not very well: I think there’s a lot of confusion and misinterpretation surrounding these terms.

What’s your favorite memory related to challenging a widely held belief?

• The thrill of presenting a well-reasoned argument that shifts people’s perspectives.
• The satisfaction of seeing my own skepticism lead to a deeper understanding of a topic.
• I tend to avoid conflict and stick with what’s comfortable.

How do you feel about the null-hypothesis significance test?

• It’s a flawed method that should be used with extreme caution.
• It’s a useful tool, but it has its limitations.
• It’s the gold standard for statistical inference.

What’s your favorite example of how the null-hypothesis significance test can be misleading?

• The case of Igor Hopewell’s dissertation, where the test could lead to the wrong conclusion about his research.
• Any example where a statistically significant result doesn’t actually have practical significance.
• I can’t think of a specific example, but I’m open to learning more about the limitations of the test.

A colleague is presenting their research findings based solely on p-values. What is your first response?

• I politely point out the limitations of relying solely on p-values and suggest considering effect sizes and confidence intervals.
• I internally cringe but keep my concerns to myself to avoid coming off as overly critical.
• I’m impressed by their rigorous statistical analysis.

You have a choice of attending a workshop on Bayesian statistics or reading another paper on the limitations of the NHD test, which do you choose?

• The Bayesian statistics workshop, because I’m eager to learn practical alternatives to the NHD test.
• The paper on NHD limitations, because I want to solidify my understanding of the issues before exploring alternatives.
• Neither, I’m happy with my current understanding of statistical analysis.

How comfortable are you with uncertainty in scientific findings?

• Very comfortable: Uncertainty is an inherent part of the scientific process.
• Somewhat comfortable: I acknowledge that not all findings are definitive.
• Uncomfortable: I prefer clear-cut answers and definitive conclusions.

What is your idea of a perfect statistical analysis?

• One that uses a Bayesian approach to estimate the probability of different hypotheses given the data.
• One that clearly outlines the limitations of the chosen statistical method.
• One that produces a statistically significant p-value.

You are at a conference and someone asks “What do you think about Rozeboom’s critique of null-hypothesis testing?”. What’s the actual answer?

• “It was a groundbreaking paper that highlighted the need for a more nuanced approach to statistical inference.”
• “I think it’s important to be aware of the limitations of null-hypothesis testing, but I still believe it can be a useful tool.”
• “I’m not really familiar with Rozeboom’s work. Can you tell me more about it?”

How prepared are you for a future where probabilistic inference is the dominant paradigm in your field?

• Very prepared: I’ve already embraced a Bayesian perspective.
• Somewhat prepared: I’m actively learning about probabilistic approaches.
• Not prepared at all: I need to catch up on this shift in thinking.

What do you think you need to fully embrace a probabilistic approach to scientific inference?

• A deeper understanding of Bayesian statistics and its applications.
• More practical experience applying probabilistic methods to real-world problems.
• A shift in mindset from decision-making to continuous refinement of beliefs.

How often do you use Bayesian analysis in your own work?

• Frequently: It’s my go-to approach for statistical inference.
• Occasionally: I use it when the situation calls for it.
• Never: I’m not yet comfortable using Bayesian methods.

How confident are you in your ability to interpret statistical findings in a nuanced way that avoids the pitfalls of the NHD method?

• Very confident: I’m aware of the biases and limitations of the NHD method and focus on effect sizes, confidence intervals, and the overall strength of evidence.
• Somewhat confident: I’m still developing my ability to critically evaluate statistical results.
• Not very confident: I tend to rely on the p-value as the main indicator of significance.

How do you handle situations where your research findings contradict a widely held belief based on the NHD method?

• I carefully consider the limitations of both my own findings and the previous research, and present my results in a nuanced way that encourages further discussion and investigation.
• I’m hesitant to challenge the status quo and might downplay my findings to avoid controversy.
• I assume my findings are flawed and dismiss them without further exploration.

Do you have a strong foundation in statistical theory, including concepts like inverse probability and fiducial probability?

• Yes, I have a solid understanding of these concepts.
• I’m familiar with the terms, but need to review the specifics.
• No, these concepts are new to me.

How well do you stick to your convictions about the limitations of the NHD test when everyone else seems to accept it without question?

• Very well: I’m willing to stand my ground and advocate for a more nuanced approach, even when it’s unpopular.
• I try to, but it’s hard to resist the pressure to conform.
• Not well at all: I tend to go along with the majority opinion.

Which of the following is most accurate when it comes to your approach to interpreting research findings?

• I prioritize understanding the practical implications of the findings over simply focusing on statistical significance.
• I try to strike a balance between statistical significance and practical relevance.
• Statistical significance is paramount – if the p-value is less than .05, the findings are important.

To what degree do you experience anxiety when your research findings contradict a widely held belief?

• Significant anxiety: I worry about the implications for my career and reputation.
• Moderate anxiety: I feel some pressure to conform, but ultimately prioritize the pursuit of truth.
• Minimal anxiety: I trust my scientific judgment and welcome intellectual debate.

Which of these best describes your current understanding of the debate surrounding the NHD test?

• I’m well-versed in the arguments for and against the NHD test and have formed my own informed opinion.
• I’m aware of the debate but haven’t delved deeply into the arguments.
• I had no idea there was a debate about the validity of the NHD test.

What is your current biggest challenge when it comes to applying more nuanced statistical methods like Bayesian inference?

• Lack of time and resources to dedicate to learning and implementing these methods.
• Difficulty understanding the complex mathematical underpinnings of these approaches.
• Resistance from colleagues and collaborators who are entrenched in the traditional NHD framework.

What’s the first thing that comes to mind when you encounter a p-value in a research paper?

• “I wonder what the effect size is and whether this finding is practically meaningful.”
• “I hope it’s less than .05!”
• “I wonder if they controlled for all the relevant confounding variables.”

How do you handle the pressure to publish research findings that are statistically significant, even if you have reservations about the methodology?

• I prioritize methodological rigor and transparency, even if it means delaying publication or choosing a less prestigious outlet.
• It’s a struggle, but I try to find a balance between publishing impactful research and maintaining my scientific integrity.
• I feel pressured to publish significant findings, even if it means compromising my research standards.

How would you describe your relationship to statistics?

• A necessary tool that I’m always working to better understand and apply in service of scientific inquiry.
• A source of frustration and confusion that I try to avoid as much as possible.
• A fascinating subject that I enjoy learning about in my free time.

Are you stuck in a cycle of relying on p-values without fully grasping the underlying concepts of statistical inference?

• No, I’m confident in my understanding of statistical inference and use p-values appropriately.
• Maybe, I sometimes question whether I’m using statistics correctly.
• Yes, I often feel lost in the world of statistics.

What would you say are your top struggles right now when it comes to critically evaluating research, particularly in regards to the use of statistical analysis?

• Identifying potential biases in the study design and data analysis that might influence the results.
• Understanding the limitations of different statistical methods and how those limitations might affect the interpretation of findings.
• Distinguishing between statistically significant findings and practically meaningful results.

What is your long-term goal in terms of your statistical literacy and ability to conduct rigorous scientific research?

• To become a thought leader in my field who champions a more nuanced and sophisticated approach to statistical analysis.
• To be able to confidently design, conduct, and interpret my own research using appropriate statistical methods.
• To be able to understand and critically evaluate the research of others.

What do you think is missing in the way statistics is typically taught in your field?

• A greater emphasis on the conceptual understanding of statistical inference, rather than just the rote application of formulas.
• More hands-on experience applying statistical methods to real-world data sets.
• A more critical perspective on the limitations of traditional statistical approaches like the NHD test.

New research emerges that supports the use of the NHD test in specific contexts. How do you respond?

• I approach the research with a healthy dose of skepticism, carefully evaluating the methodology and the strength of the evidence.
• I’m relieved to have my faith in the NHD test restored.
• I dismiss the research without giving it much consideration.

Someone asks, “How are you doing with staying current on the latest developments in statistical inference?” What’s the actual answer?

• “It’s definitely a challenge to keep up with everything, but I’m actively trying to learn about new approaches like Bayesian analysis.”
• “I’m a bit behind, but I’m planning to catch up on the latest research soon.”
• “To be honest, I haven’t really thought much about it lately.”

What’s your go-to resource for staying up-to-date on best practices in statistical analysis?

• Academic journals that specialize in statistical methodology.
• Online courses and workshops taught by experts in the field.
• I rely on my colleagues to keep me informed.

What statistical concept do you most want to explore and learn about in more depth?

• Bayesian statistics and its potential applications in my field of study.
• The philosophy of statistics and the role of probability in scientific inference.
• Advanced statistical modeling techniques that can account for complex data structures.

What’s your idea of a perfect world when it comes to statistical literacy among researchers?

• All researchers have a strong conceptual understanding of statistical inference and can critically evaluate the appropriateness of different statistical methods.
• Statistical analysis is seen as a tool for uncovering truth, rather than just a means to an end (i.e., getting published).
• Researchers prioritize transparency and reproducibility in their statistical analyses.

What is your strongest asset when it comes to navigating the complexities of statistical analysis?

• My willingness to admit when I don’t know something and my eagerness to learn from others.
• My critical thinking skills and ability to evaluate evidence objectively.
• My perseverance and determination to master challenging concepts.