This week covers two studies. The first is a fascinating paper published recently by the World Bank, which suggests features of language are linked to gender disparities within and between countries. The second is my own — I handed my thesis in a week ago, which goes some way to explaining why this edition of Interesting Things is so late. I certainly don't mean to presume you, loyal reader, have any level of interest in the inner workings of New Zealand digital government initiatives, but a few of you have been naive enough to ask me about my research, so I'm indulging myself.
Speaking of indulgent: before we get into it, I'd like to note this is the first edition since the first birthday of Interesting Things I Come Across. The newsletter has grown substantially since it started as a way to keep in touch with a few BCG colleagues — there are now hundreds of you, many of whom I've never met. Thanks to those who reply with more interesting points, and to those who forward on to friends. I aspired to write weekly, and for the first year I've managed one edition every eleven days. For the next year, I'm aiming for fortnightly editions. As always, your thoughts and correspondence are welcome.
Talking about gender
The main claim of this World Bank paper is that "gender languages appear to reduce women's labor force participation and perpetuate support for unequal treatment of women." Let's start at the beginning of the abstract:
Languages use different systems for classifying nouns. Gender languages assign many — sometimes all — nouns to distinct sex-based categories, masculine and feminine.
For example, French nouns are prefaced with le (masculine) or la (feminine). While English has a few gendered — actor/actress, for example — there are no gendered classifiers. Gendered classifiers affect how people understand and describe inanimate objects. For example, the German word for bridge, Brücke, is a feminine noun, while the Spanish word, puente, is a masculine noun. In another study, Germans and Spaniards were asked to describe photos of bridges. Germans used words like 'beautiful', 'elegant', 'fragile', 'peaceful', 'pretty', and 'slender', while Spaniards used words like 'big', 'dangerous', 'long', 'strong', 'sturdy', and 'towering'. (thanks Wikipedia). But back to the World Bank abstract — gendered classifiers are linked to far more than adjective use:
Drawing on a broad range of historical and linguistic sources, this paper constructs a measure of the proportion of each country's population whose native language is a gender language. At the cross-country level, this paper documents a robust negative relationship between the prevalence of gender languages and women's labor force participation. It also shows that traditional views of gender roles are more common in countries with more native speakers of gender languages.
Correlation is not causation, of course, but the paper does control for features within and between countries and finds the patterns are statistically robust even when accounting for national development, and "unobservable" aspects of culture.
This adds to a lot of hard evidence showing language constructs the world we experience — even the colours we are able to percieve. Readers of Interesting Things will know that I view the world in terms of systems and structures, and language is a perfect example of a system we don't have much control over. We aren't able to choose whether we grow up natively speaking a gendered language or not, yet this appears linked to how society works — or doesn't — for women. The World Bank infamously demands structural adjustments in struggling economies — what would language reform look like?
Why does the New Zealand government use algorithms to augment decision-making?
In short, my research looks at why senior bureaucrats deploy digital tools to the front lines of public service delivery: what motivates managers to use algorithms in decision-making? I've copy-pasted the whole abstract below in case you're interested, and am happy to discuss further or send a copy of the whole thing if you're feeling particularly insomnolent (a real word!).
This paper explores the motivations of public sector managers in developing and deploying digital tools to support decision-making at the front-lines of public service delivery.
In traditional conceptions of bureaucracy, managers set rules and enforce compliance down the hierarchy. In 1980, Michael Lipsky flipped this view by suggesting the actions of ‘street-level bureaucrats’, through their autonomic exercise of discretion, in fact define bureaucracy. Yet recent technological advancements provide managers new means to enforce hierarchy by controlling street-level bureaucratic behaviour: for example, managers can deploy digital decision support tools that limit human involvement by automating part or all of decision-making processes. In response, public administration scholars have often viewed digital tools as a threat to street-level discretion, and in turn, street-level bureaucracy theory as a whole.
This research draws primarily on semi-structured interviews with senior managers responsible for the implementation of two digital decision support tools used in New Zealand’s Ministry of Social Development. Results provide empirical evidence that public sector managers deploy digital tools not to curtail, but to support street-level bureaucrats’ discretion. Managers appear to be motivated not by increased control over front-line staff, but rather, by improving clients’ experience of the system and decreasing long-term service costs. These motivations are consistent with the Digital Era Governance paradigm of public administration and also suggest rear-guard elements of New Public Management remain. In providing empirical examples of managers’ intentions to use digital decision support tools to support discretion, these findings re-affirm street-level bureaucracy theory, and suggest digital tools can provide managers with front-line insights that reduce the need for prescriptive rules and processes.
|
|
|
|
