Are We Over Statistics?

April 11, 2016

by — Posted in Peer Edited Submissions

Statistics is happening – at #statistics + #college; with #exam + #stats; amid #pedagogy + #statistics – statistics is mingling with university students, embedding itself within education as late-night paper marathons, t-test tutoring, lengthy lectures, and, crucially, with questions of relevance. As big data becomes as familiar as air[1], and numbers as common as conversation, students are engaging in discussion around the ‘why’ of statistical education. Questions of the purpose and importance of statistics education are debated with increasing (and often frustrated) urgency by our students: why, in child, youth, and family studies, where critical, politicized practice nourishes our work[2], do we need to learn about normal distributions? Why, when statistical sciences are so decisively implicated in historical[3] and ongoing colonization[4], do we need to dedicate so much time to t-tests? Why is my statistics class not able to ask critical questions of quantitative analysis?


As a feminist educator, I am inspired by our students’ questions: Why do we need statistical education in child and youth care? I refuse to root our students’ provocations in any socially sanctioned malaise in the humanities, where “I hate math” becomes a complacent refrain. Rather, questions of statistics pedagogies in our field are vibrant. That they are grumbled with such persistence is a reminder, as per Sarah Ahmed[5], that “our feminist political hopes rest with over-sensitive students. Over-sensitive can be translated as: sensitive to that which is not over.” Statistics is not over in child and youth care. Our students are not over quantitative data analysis. Statistics is not over (for) children, youth, and families[6]. As child and youth care educators, we cannot succumb to hegemonic legacies of politically neutral statistics. Instead, we must engage complex questions of pedagogy, data, and analysis as we confront our nuanced, and vulnerable relationship with statistics.

Are We Over Pedagogies of Quantitative Analysis?

Quantitative data analysis is not as firm in epistemological hegemony as dominant Euro-Western institutions might have us believe. We must ask, are statistics already a collective[7], an experiment, or a provocation; a moment just as entangled with life as any other. When Elizabeth Wilson calls us to “turn with more curiosity to the debris that our [disciplinary] politics have generated”[8], I feel the power-laden histories of quantitative research boldly foreground the everyday political gestures that sanitize statistics in the academy, centering these practices in our classrooms and reminding us that invisibility is never an accident. As Bruno Latour[9] and John Law[10] burrow into academic hinterlands, making public the structures that require we “eat [our] epistemological greens”[11] to be afforded the academic validity of certain knowledge-producing practices, I understand that statistics is a tradition, an assemblage of practices and consequences, but not an ontological condition. When Isabelle Stengers demands that we slow down, “that we don’t consider ourselves authorized to believe we possess the meaning of what we know”[12], I am reminded that statistics is already fractured and bifurcating – and that the possibilities for critical statistics meet these voids with hope. As Jordan Ellenberg advocates for ‘number sentences’ in mathematical education as a “radical act of insisting that mathematics has meaning”[13], I am reminded that numbers are never far from the fleshy humans who craft them and that digits are infused with the same political desire, uncertainties, and possibilities[14] as our work with children, youth, and families. When Nate Silver and fivethirtyeight wretch statistics from the façade of objectivity, making ethical the role of statistics in ongoing systemic racism, colonization, and minoritization[15], I know that statistics are quietly overflowing with the very real, consequential analyses that matter[16] in child and youth care.

When data analysis becomes about abstracting the complicated bodied activities of children and families to tables and figures, and as our government remembers murdered and missing Indigenous women and girls as numbers as if digitizing suffering somehow absolves settlers of responsibility for immense gender-based violence[17], we are forced to confront a statistics that is deeply vulnerable and always knotted within the impassioned, care-ful, political tangles[18] that animate child and youth care. The challenge of statistics pedagogies is that of meeting with a statistics that is exposed through its own power; that demands accountability, critique, and creativity; and an approach to engaging quantitative analysis that refuses to play by the rules of objectivity, as it demands as much political and reflexive engagement from our students as any other exercise.

Are We Over Quantitative Data?

 In pouring over the lengthy debates around pedagogies of quantitative data analysis[19], I am struck by the consistency of two concepts: data and analysis. While pedagogues rooted in this debate question how we might reconceptualize our undergraduate statistics curriculums[20], the ‘fact’ of data[21] is not up for debate, nor is the requirement that analysis will happen[22]. If we, as instructors orient our pedagogies towards statistics that are vulnerable, fragmented, and accountable, both ‘data’ and ‘analysis’ are fabricated in frictions. These frictions do not demand that we abandon teaching statistics – in fact, I believe that they highlight the urgency of engaging with theories of statistics.

Our pedagogies of statistics demand a vastly different practice if we consider data to be always infused with subjectivity, uncertainty, and the potential for error. If we figure data as “passive objects, waiting to be coded or granted shape and significance through the interpretative work of researchers”[23], teaching statistics requires a different ethical framework than if we believe data to be “data (under erasure), data-undone, data-rethought, data-particles, or maybe data-becoming”[24]. If we take the nature and consequences of data seriously, what matters is that we invite questions of what data is and does to interject in our pedagogies: What happens if we put the nature of data at stake, such that we might wonder how each collection of quantitative data is distributed differently not only in count but also in kind? How can we wonder with our students if data might be “more about responsibility than epistemology”[25]? How can we teach the fundamentals of data analysis without dissolving the contested realities of the data under analysis?

Are We Over Quantitative Data Analysis?

 When we attend to the critical character of data, we must also put at risk our conceptions of what it might mean to ‘analyze’ data in our courses. When the very nature of what ‘counts’ as ‘rigorous’ quantitative analysis is up for debate in the academy writ large[26], how instructors confront the challenges of constructing an ethical and practical balance between teaching the concepts necessary for statistics literacy while also working with our students to cultivate a critical quantitative consciousness is deeply consequential. What if we question how we might bring hypothesis testing into our teaching in a such a way that calculating the correct p-value[27] is just as important as understanding the debates surrounding the efficacy of that p-value[28]? When error bars[29], p-hacking[30], and ‘Repligate’[31] are emblematic of the politics of reproducibility and reliability[32], open data and transparency[33], and post-publication peer-review[34] in quantitative research, might questions of how we do statistics be just as consequential as the questions we create to test if students can do statistics?

Are We Over Statistics in Child and Youth Care?

Statistics is not over in child and youth care. Quantitative analysis is not impenetrable. Data is not a singularity. Analysis is not one practice. Statistics matter. I have unending faith that our students and statistics are not too critical for one another. I wonder how our students might risk caring for, or loving (with), the critically high stakes[35] that populate our means, medians, and modes? How can we hospice the violent, colonial and capitalist projects of quantitative analysis[36] while also holding any potential revolutionary force in/of numbers? What might never erasing the bodies, the moments, and the relations that compose data and analysis do to our practices of t-tests, correlations, and inference-making?

Engaging the critical questions that care-ful, contingent, and contested relationships between statistics and our students generate requires that we put at risk the pedagogical traditions of quantitative analysis in the academy, while holding that there is value and vitality in statistical education. Teaching statistics becomes a practice of collectively navigating[37] the messiness of data and analysis. Just as we teach statistical theories, we must support critical, creative, and accountable engagement with quantitative analysis. In recognizing that it is imperative that our courses are able to ask critical questions of quantitative analysis, we acknowledge that statistics is an ongoing project – concerned with how we might reclaim the power in/of data and analysis and mount the critical, collective, ethical, and responsive pedagogies of statistical data analysis demanded by our students.

[1] Baker, L. (2015). Welcome to the statistics revolution. Retrieved from Big Data.

[2] Skott-Myhre, H., Pacini-Ketchabaw, V., & Skott-Myhre, K. (2015). Immanent approaches and liminal encounters in youth work, early education, and psychology. Retrieved from IJCYFS Review.

[3] Walter, M., & Andersen, C. (2013). Indigenous statistics: A quantitative research methodology. New York, NY: Routledge.

[4] Statistics Canada. (2013). Aboriginal peoples in Canada: First Nations People, Metis, and Inuit. Retrieved from Statistics Canada.

[5] Ahmed, S. (2015). Against students. Retrieved from Feminist Killjoy.

[6] British Columbia Representative for Children and Youth. (n.d.). Reports & publications. Retrieved from Office of the Representative of Children and Youth.

[7] Keller, S.A. (2010). Vital statistics. Nature, 467(7318).

[8] Wilson, E.A. (2015). Gut feminism. Durham, NC: Duke University Press. [Citation from page 43]

[9] Latour, B. (2005). Reassembling the social: An introduction to actor-network-theory. Oxford, England: Oxford University Press.

[10] Law, J. (2004). After method: Mess in social science research. New York, NY: Routledge.

[11] Law, J. (2006). Making a mess of method. Retrieved from [Citation from page 2]

[12] Stengers, I. (2005). The cosmopolitical proposal. In B. Latour & P. Weibel (Eds.), Making things public: Atmospheres of democracy (pp. 994–1004). Cambridge, MA: MIT Press. [Citation from page 995]

[13] Ellenberg, J. (2014). Number sentences: Stephen Colbert thinks they’re silly. They’re not. Retrieved from Slate.

[14] White, J. (2015). An ethos for the times: Difference, imagination, and the unknown future in child and youth care. International Journal of Child, Youth and Family Studies, 6(4), 498-515.

[15] Avirgan, J. (2016). Bad data – and worse decisions – poisoned Flint. Retrieved from FiveThirtyEight.

[16] de Finney, S., Dean, M., Loiselle, E., & Saraceno, J. (2011). All children are equal, but some are more equal than others: Minoritizational, structural inequities, and social justice praxis in residential care. International Journal of Child, Youth and Family Studies, 2(3&4), 361-384.

[17] Coon, E. (2015). tewahtatowi: We carry ourselves. Retrieved from IJCYFS Review.

[18] White, J. (2015). Thinking about borders and tangled meshes. Retrieved from IJCYFS Review.

[19] Horton, N. (2016). Response collective to “Mere Renovation is Too Little Too Late: We Need to Rethink Our Undergraduate Curriculum from the Ground Up (Cobb, 2015)”. Retrieved from

[20] Cobb, G.W. (2015). Mere renovation is too little too late: We need to rethink our undergraduate curriculum from the ground up. American Statistician, 69(4).

[21] Gould, R. (n.d.). Augmenting the vocabulary used to describe data. Retrieved from

[22] Temple Lang, D. (n.d.). Authentic data analysis experience. Retrieved from

[23] Koro-Ljungberg, M., & MacLure, M. (2013). Provocations, re-un-visions, death, and other possibilities of “data”. Cultural Studies – Critical Methodologies, 13(4), 219-222. [Citation from page 219]

[24] Ibid.

[25] Lather, P. (2010). Engaging science policy: From the side of the messy. New York, NY: Peter Lang. [Citation from page 19]

[26] Lakens, D. (2015). On the challenges of drawing conclusions from p-values just below 0.05. PeerJ, 3(e1142).

[27] Colquhoun, D. (2014). An investigation of the false discovery rate and the misinterpretation of p-values. Royal Society Open Science.

[28] Rhodes, E. (2015). Liberating, or locking away, our best tools? Retrieved from The British Psychological Society: The Psychologist.

[29] Vaux, D.L. (2012). Research methods: Know when your numbers are significant. Nature, 492(7428), 180-181.

[30] Nuzzo, R. (2014). Statistical method: Statistical errors. Nature, 506(7487).

[31] Chambers, C. (2014). Physics envy: Do ‘hard’ sciences hold the solution to the replication crisis in psychology? Retrieved from The Guardian.

[32] The Academy of Medical Sciences. (2015). Reproducibility and reliability of biomedical research. Retrieved from AcMedSci.

[33] Lewandowsky, S., & Bishop, D. (2016). Research integrity: Don’t let transparency damage science. Nature, 529(7587).

[34] Barr, D.J. (2014). Anatomy of a statistical artifact: Eudaimonic well-being and geonomics. Retrieved from Datahowler.

[35] Skott-Myhre, K., & Skott-Myhre, H. (2015). Revolutionary love: CYC and the importance of reclaiming our desire. International Journal of Child, Youth and Family Studies, 6(4), 581-594.

[36] de Oliveira Andreotti, V., Stein, S., Ahenakew, C., Hunt, D. (2015). Mapping interpretations of decolonization in the context of higher education. Decolonization: Indigeneity, Education & Society, 4(1), 21-40.

[37] Silberzahn, R., et al. (n.d.). Many analysts, one dataset: Making transparent how variations in analytical choices affect results. Open Science Framework.

This article was peer-edited by Jeffrey Ansloos

Image used with permission from Dear Data – Week 49: A Week of “Data”

About Nicole Land

Nicole Land is a PhD student, sessional instructor, and research assistant in the School of Child and Youth Care at the University of Victoria, where she also works with early childhood educators as a pedagogical facilitator. Drawing on feminist science studies and material feminist perspectives, she focuses on re-imagining how fat, muscles, and movement can be practiced, move, and matter in work with young children. Nicole is currently preparing to teach an undergraduate data analysis course for the second time.
Share on FacebookEmail this to someoneTweet about this on TwitterPrint this page

Leave a Reply

Your email address will not be published. Required fields are marked *