Quotessence
Home / Topics / Data Analysis Quotes

Data Analysis Quotes

Browse 54 quotes about Data Analysis.

Data Analysis Quotes

“Blogging, writing conventional articles, and being science consultant and pocket protector ninja to various web portals and TV programs, quite often trying to promote the penicillin of hard data to people who had no interest in being cured of their ignorance.”

“G. Stanley Hall, a creature of his times, believed strongly that adolescence was determined – a fixed feature of human development that could be explained and accounted for in scientific fashion. To make his case, he relied on Haeckel's faulty recapitulation idea, Lombroso's faulty phrenology-inspired theories of crime, a plethora of anecdotes and one-sided interpretations of data. Given the issues, theories, standards and data-handling methods of his day, he did a superb job. But when you take away the shoddy theories, put the anecdotes in their place, and look for alternate explanations of the data, the bronze statue tumbles hard. I have no doubt that many of the street teens of Hall's time were suffering or insufferable, but it's a serious mistake to develop a timeless, universal theory of human nature around the peculiarities of the people of one's own time and place.”

“Heightened data competence can therefore ensure data is used to improve the lives and experiences of LGBTQ people rather than only serve the interests of, what Catherine D’Ignazio and Lauren F. Klein described as, the three S's: science (universities), surveillance (governments), and selling (corporations).”

“Hacking described his research interest ‘in classifications of people, in how they affect the people classified, and how the affects on the people in turn change the classifications.’ Hacking labeled the subjects of these studies ‘moving targets’ because researchers’ investigatory efforts change them in ways so ‘they are not quite the same kind of people as before.”

“The cleaning of data can remove its queerness: paper surveys where respondents score out the response options ‘female’ and ‘male’ and write their own answer, interview recordings were participants flip the focus and ask questions of the researcher, census returns where LGBTQ couples identify themselves as ‘married’ even when governments do not recognize same sex marriage. These examples demonstrate how collection methods can fail to restrict how participants share data about their lives and experiences. … cleaning, which involves the removal of data that breaks established rules”

“More information means less ignorance and a greater chance of rational and better decisions and not those based on illusions, hope, preconceived notions or perceptions. The danger from so much data—there is no definition of what is optimum—is that there are chances of overanalysis or falling into a conspiracy theory trap.”

“Do you need more data? Do not assume the need for more data -- enough evidence of a problem might already exist to justify the need for action. Also explore who is already engaged in data practices on the topic to see if resources could support existing initiatives rather than create something afresh. The collection, analysis and use of data are resource-intensive. Before work begins, you therefore need to ask if this is the best use of time, resources and energy to address injustices that face LGBTQ people.”

“Do you elevate LGBTQ lives and critically examine the invisibility of majority characteristics? One of data’s strengths is its power to tell stories, which can shifts hearts and minds and encourage others to take action. However, increased visibility alone is not enough. A queer approach also problematizes the distinction between the center and the margins so the invisibility of majority identity characteristics, such as cis and heterosexual, are brought into focus and critically examined.”

“Are your ways of working open, accessible and transparent? Traditional approaches to quantitative data collection and analysis are misunderstood as an objective account of reality; an assumption that masks decisions made throughout the design process. A queer approach to data is also influenced by biases and assumptions; those engaged in queer data practices therefore need to describe how decisions are made, in accessible language, and its effect on the results presented. Openness about the limitations of data helps ensure that an undercount or misrepresentation of data about LGBTQ people is not used undermine political and social advances.”

“Are damaging data practices and systems capable of reform? Re-evaluate your relationship to data and assess whether existing practices and systems are capable of reform. If reform seems possible, question who is best placed undertake this work. When reform fails, or efforts to reform risk keeping a damaging system alive for longer, consider if an abolitionist approach might put data in the hands of those most in need.”