class: center, middle, inverse, title-slide # Community Features and School Achievement Gaps ## And some other stuff ### Daniel Anderson ### June 22, 2018 --- # Introduction * Much recent focus on open data in research generally * Open data tend to be rare in educational research + Privacy concerns -- .Large[.bolder[.center[NCLB Required Publicly Available Data]]] -- * School-level data * Percent proficent in each of at least four proficiency categories * Disaggregated by student subgroups --- # Illustrating one approach * Simulate some data from two distributions + Group 1: `\(n = 200\)`, `\(\mu = 200\)`, `\(\sigma = 10\)` + Group 2: `\(n = 500\)`, `\(\mu = 210\)`, `\(\sigma = 8\)` * Cut Scores: 190, 205, 215 .gray[.tiny[(totally made up)]] <img src="sicss-pres_files/figure-html/sim-1.png" style="display: block; margin: auto;" /> --- # Reardon & Ho Method<sup>.gray[.tiny[(2)]]</sup> * Calculate the empirical CDF of each distribution * Pair the ECDFs * Calculate the area under the paired curve * Transform it to an effect-size measure (standard deviation units) .footnote[(2) Reardon, S. F., & Ho, A. D. [(2015)](http://journals.sagepub.com/doi/abs/10.3102/1076998615570944). Practical issues in estimating achievement gaps from coarsened data. *Journal of Educational and Behavioral Statistics*, *40*, 158–189.] -- .pull-left[ <!-- --> ] -- .pull-right[ <!-- --> ] --- # Transformation to effect size .Large[ $$ V = \sqrt{2}\Phi^{-1}(AUC) $$ ] * `\(V\)` = Cohen's `\(d\)` under the assumption of respective normality * In this case, Cohen's `\(d\)` = -1.07 and `\(V\)` = -1.06. --- # Why does this all matter? <img src="sicss-pres_files/figure-html/props-1.png" style="display: block; margin: auto;" /> --- class: inverse # Purpose -- * Evaluate `\(V\)` with empirical data -- + Compare estimates from full data and manually coarsened data. Compare results to Cohen'd `\(d\)` estimated with full data. -- * Apply these methods to publicly available data to investigate between-school differences in achievement gaps -- + Specifically using geo-spatial mapping, overlaying data about the surrounding area --- class: inverse center middle background-image: url(anderson_pres_ncme18_files/img/sky.jpg) background-size: cover # Results --- # Comparing `\(V_c\)` and `\(d\)` Both continuous: `\(r = 0.87/0.86\)`.  --- .pull-left[ ### Comparing `\(V_d\)` and `\(d\)` .center[ `\(r = 0.73/0.72\)` ] <img src = "./anderson_pres_ncme18_files/img/vd_bivariate_discrete.png" height = 450/> ] .pull-right[ ### Comparing `\(V_c\)` and `\(V_d\)` .center[ `\(r = 0.83/0.89\)` ] <img src = "./anderson_pres_ncme18_files/img/vv_bivariate.png" height = 450/> ] --- ## Distribution of differences: `\(V_c\)` and `\(d\)` .pull-left[ .center[ `\(\mu = -0.12, \sigma = 0.15\)` ] ] .pull-right[ .left[ `\(\mu = -0.16, \sigma = 0.16\)` ] ] <img src = "./anderson_pres_ncme18_files/img/vd_distributions_continuous.png" height = 375/> --- .pull-left[ ### Comparing `\(V_d\)` and `\(d\)` .center[ `\(\mu = -0.10/-0.13, \sigma = 0.21/0.23\)` ] <img src = "./anderson_pres_ncme18_files/img/vd_distributions_discrete.png" height = 425/> ] .pull-right[ ### Comparing `\(V_d\)` and `\(V_c\)` .center[ `\(\mu = 0.02/0.03, \sigma = 0.17/0.15\)` ] <img src = "./anderson_pres_ncme18_files/img/vv_distributions.png" height = 425/> ] --- class: inverse center background-image: url(anderson_pres_ncme18_files/img/map.jpg) background-size: cover # Substantive Investigations --- # Achievement Gap Distributions .grey[Reminder: School-level Distributions] <center><img src = "anderson_pres_ncme18_files/img/school_achievement_gaps.png" height = 425/></center> --- # Alameda County ### Median Housing Cost <iframe seamless src="anderson_pres_ncme18_files/maps/alameda_housing.html" width="100%" height="425"></iframe> --- # Alameda County ### `\(n\)` Identifying as Black <iframe seamless src="anderson_pres_ncme18_files/maps/alameda_black.html" width="100%" height="425"></iframe> --- # Alameda County ### `\(n\)` Income/Poverty Ratio > 2.0 <iframe seamless src="anderson_pres_ncme18_files/maps/alameda_poverty.html" width="100%" height="425"></iframe> --- # Conclusions * Effect size appeared well estimated from coarsened data + `\(V\)` was similar to Cohen's `\(d\)` with these empirical data * The vast majority, but not all, schools had sizeable estimated achievement gaps * Clear geographic clustering of achievement gaps was evident + Did Not appear to depend on the Census data investigated here --- class: inverse right background-image: url(anderson_pres_ncme18_files/img/quickly.jpg) background-size: cover # Really quick: These slides! --- # Reproducibility & Transparency * I care a lot about it * Slides are pretty (I think) * Super customizable (css) * Open source, transparent * [slidex](https://github.com/datalorax/slidex) helps you transition from Microsoft Powerpoint! --- class: inverse background-image: url(https://github.com/datalorax/slidex/raw/master/docs/slidex-preview.gif) background-size: contain --- class: inverse left background-image: url(anderson_pres_ncme18_files/img/thanks.jpg) background-size: cover # Thanks so much! [
datalorax](https://github.com/datalorax) [
datalorax_](https://twitter.com/datalorax_) [
datalorax.com](http://www.datalorax.com/) [
daniela@uoregon.edu](mailto:daniela@uoregon.edu) <br/> <br/> <br/> <br/> <br/> .Large[ .center[ Slides available at [datalorax.com/talks/sicss-seattle](http://www.datalorax.com/talks/sicss-seattle) ] ]