All Items 25

Teaching the next generation of statistics students to "think with data": Special issue on statistics and the undergraduate curriculum

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Teaching the next generation of statistics students to "think with data": Special issue on statistics and the undergraduate curriculum
Ensuring that mathematics is relevant in a world of data science

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

The recent growth of data science has been remarkable. Analysts now have rich data and powerful computational tools to help answer important questions. Examples of ways that insights can be wrangled from this information abound in diverse areas. This has led some to dub computational thinking (or fluency) as the "new literacy" on par with writing and quantitative skills. A major unanswered question relates to the role of mathematics in the training of future data scientists. How can we be sure that data science is on a firm mathematical and statistical foundation? In the article, we will consider what courses in mathematics would best prepare future data scientists.

Ensuring that mathematics is relevant in a world of data science
Distortion product otoacoustic emissions and intracranial pressure during CSF infusion testing

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Background: A noninvasive method to monitor changes in intracranial pressure (ICP) is required for astronauts on long-duration spaceflight who are at risk of developing the Visual Impairment/Intracranial Pressure syndrome that has some, but not all of the features of idiopathic intracranial hypertension. We assessed the validity of distortion product otoacoustic emissions (DPOAEs) to detect changes in ICP. Methods: Subjects were eight patients undergoing medically necessary diagnostic cerebrospinal fluid (CSF) infusion testing for hydrocephalus. DPOAE measurements were obtained with an FDA-approved system at baseline and six controlled ICP levels in ∼3 mmHg increments in random order, with a range from 10.8 ± 2.9 mmHg (SD) at baseline to 32.3 ± 4.1 mmHg (SD) at level 6. Results: For f2 frequencies between 800 and 1700 Hz, when ICP was ≥ 12 mmHg above baseline ICP, DPOAE angles increased significantly and DPOAE magnitudes decreased significantly, but less robustly. Discussion: Significant changes in DPOAE angle and magnitude are seen when ICP is ≥ 12 mmHg above a subject's supine baseline ICP during CSF infusion testing. These results suggest that the changes in DPOAE angle and magnitude seen with change in ICP are physiologically based, and suggest that it should be possible to detect pathological ICP elevation using DPOAE measurements. To use DPOAE for noninvasive estimation of ICP during spaceflight will require baseline measurements in the head-up, supine, and head-down positions to obtain baseline DPOAE values at different ICP ranges

Distortion product otoacoustic emissions and intracranial pressure during CSF infusion testing
Making progress in a crowded market

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

In their wide-ranging, insightful and provocative paper, Ograjenšek Gal (2016) restate and enunciate our shared goal to help motivate students to learn to use data to provide answers to real-world problems. They also provide specific suggestions about additional ways to create more opportunities for scaffolding and practice of qualitative thinking in our courses and curriculum. I commend them for this guidance—which is wholly consistent with the sentiments of many—including Brown Kass (2009), the INGEnIOuS project, http://www.ingeniousmathstat.org and Cobb (2015). All suggest a broad definition of statistics, encourage efforts to ensure that statistics is a vibrant choice for students and call for major (radical?) changes to our curriculum to support this shift. While there have been major efforts in recent years to adapt our curriculum to provide exposure to the excitement of statistics [see for example several of the rich datasets in Gould, (2010)], students need to be more directly engaged with a broad range of new data-related topics to be successful in our increasingly diverse discipline (Horton, 2016; Ridgway, 2015). The growth of data science poses many challenges and opportunities. As the authors describe, to be relevant in this world awash in data, statistics students need exposure to quantitative and qualitative approaches for the planning and design of studies, conduct and analysis, as well as interpretation and communication of results and implications. I will briefly consider the implications of this paper as well as barriers for implementation of the approaches enunciated in this paper in two distinct areas: (1) our introductory courses as well as (2) our undergraduate programs in statistics.

Making progress in a crowded market
Reflectance measures from infant ears with normal hearing and transient conductive hearing loss

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Objective: The objective is to develop methods to utilize newborn reflectance measures for the identification of middle-ear transient conditions (e.g., middle-ear fluid) during the newborn period and ultimately during the first few months of life. Transient middle-ear conditions are a suspected source of failure to pass a newborn hearing screening. The ability to identify a conductive loss during the screening procedure could enable the referred ear to be either (1) cleared of a middle-ear condition and recommended for more extensive hearing assessment as soon as possible, or (2) suspected of a transient middle-ear condition, and if desired, be rescreened before more extensive hearing assessment. Design: Reflectance measurements are reported from full-term, healthy, newborn babies in which one ear referred and one ear passed an initial auditory brainstem response newborn hearing screening and a subsequent distortion product otoacoustic emission screening on the same day. These same subjects returned for a detailed follow-up evaluation at age 1 month (range 14 to 35 days). In total, measurements were made on 30 subjects who had a unilateral refer near birth (during their first 2 days of life) and bilateral normal hearing at follow-up (about 1 month old). Three specific comparisons were made: (1) Association of ear's state with power reflectance near birth (referred versus passed ear), (2) Changes in power reflectance of normal ears between newborn and 1 month old (maturation effects), and (3) Association of ear's newborn state (referred versus passed) with ear's power reflectance at 1 month. In addition to these measurements, a set of preliminary data selection criteria were developed to ensure that analyzed data were not corrupted by acoustic leaks and other measurement problems. Results: Within 2 days of birth, the power reflectance measured in newborn ears with transient middle-ear conditions (referred newborn hearing screening and passed hearing assessment at age 1 month) was significantly greater than power reflectance on newborn ears that passed the newborn hearing screening across all frequencies (500 to 6000 Hz). Changes in power reflectance in normal ears from newborn to 1 month appear in approximately the 2000 to 5000 Hz range but are not present at other frequencies. The power reflectance at age 1 month does not depend significantly on the ear's state near birth (refer or pass hearing screening) for frequencies above 700 Hz; there might be small differences at lower frequencies. Conclusions: Power reflectance measurements are significantly different for ears that pass newborn hearing screening and ears that refer with middle-ear transient conditions. At age 1 month, about 90% of ears that referred at birth passed an auditory brainstem response hearing evaluation; within these ears the power reflectance at 1 month did not differ between the ear that initially referred at birth and the ear that passed the hearing screening at birth for frequencies above 700 Hz. This study also proposes a preliminary set of criteria for determining when reflectance measures on young babies are corrupted by acoustic leaks, probes against the ear canal, or other measurement problems. Specifically proposed are "data selection criteria" that depend on the power reflectance, impedance magnitude, and impedance angle. Additional data collected in the future are needed to improve and test these proposed criteria.

Reflectance measures from infant ears with normal hearing and transient conductive hearing loss
Male eating disorder symptom patterns and health correlates from 13 to 26 years of age

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Objective: Research on the manifestations and health correlates of eating disorder symptoms among males is lacking. This study identified patterns of appearance concerns and eating disorder behaviors from adolescence through young adulthood and their health correlates. Method: Participants were 7,067 males from the prospective Growing Up Today Study. Surveys from 1999 to 2007 (spanning ages 13−26 years) provided repeated measures data on muscularity and leanness concerns, eating disorder behaviors (purging, overeating, binge eating, use of muscle-building products), and health correlates (obesity, non-marijuana drug use, binge drinking, and depressive symptoms). Results: Latent class analyses of observations at ages 13 to 15, 16 to 18, 19 to 22, and 23 to 26 years identified 1 large Asymptomatic class and 4 symptomatic patterns: Body Image Disturbance (high appearance concerns, low eating disorder behaviors; 1.0%−6.0% per age period); Binge Eating/Purging (binge eating and purging, use of muscle-building products, low appearance concerns; 0.1%−2.5%); Mostly Asymptomatic (low levels of muscularity concern, product use, and overeating; 3.5%−5.0%); and Muscularity Concerns (high muscularity concerns and use of products; 0.6%−1.0%). The Body Image Disturbance class was associated with high estimated prevalence of depressive symptoms. Males in the Binge Eating/Purging and Muscularity Concerns classes had high prevalence of binge drinking and drug use. Despite exhibiting modestly greater appearance concerns and eating disorder behaviors than the Asymptomatic class, being in the Mostly Asymptomatic class was prospectively associated with adverse health outcomes. Conclusion: Results underscore the importance of measuring concerns about leanness, muscularity, and use of muscle-building products when assessing eating disorder presentations among males in research and clinical settings.

Male eating disorder symptom patterns and health correlates from 13 to 26 years of age
A program aimed toward inclusive excellence for underrepresented undergraduate women in the sciences

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Created to foster inclusive excellence, Smith College’s Achieving Excellence in Mathematics, Engineering, and Science (AEMES) Scholars program provides early faculty-mentored research opportunities and other programming as a way to foster success in academic outcomes for underrepresented women in science. Using academic record data, we compared Scholars’ outcomes over time with those of underrepresented students before program launch and to relevant peer comparison groups. Since its launch, AEMES Scholars have achieved significantly higher gateway life sciences course grade point averages (GPAs), rates of persistence in life and natural sciences, and participation in natural sciences advanced research relative to baseline. Gains for Scholars in gateway course GPA eliminated the significant gap that previously existed between science, technology, engineering, and mathematics (STEM)-underrepresented and other students, whereas gains in natural sciences persistence now has Scholars continuing in STEM at significantly higher rates than all other students. Many of the gains for AEMES Scholars were echoed in findings of improved outcomes for our STEM students overall since AEMES’ launch. Underrepresented students who were not part of the Scholars program also evidenced increased gateway course GPA over this same period. We discuss potential explanations for these outcomes and ongoing work aimed at achieving further inclusive excellence for women in the sciences.

A program aimed toward inclusive excellence for underrepresented undergraduate women in the sciences
Family functioning and quality of parent-adolescent relationship: cross-sectional associations with adolescent weight-related behaviors and weight status

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Background: Little is known about how factors within the general family environment are associated with weight and related behaviors among adolescents/young adults. Methods: We studied 3768 females and 2614 males, 14–24 years old in 2011, participating in the Growing Up Today Study 2. We used generalized mixed models to examine cross-sectional associations of family functioning and quality of mother- and father-adolescent relationship with adolescent/young adult weight status, disordered eating, intake of fast food and sugar-sweetened beverages, screen time, physical activity, and sleep duration. In all models, we included participant’s age and family structure. Results: Eighty percent of participants reported high family functioning and 60 % and 50 % of participants reported high-quality mother and father relationship, respectively. Among both males and females, high family functioning was associated with lower odds of disordered eating (adjusted odds ratio [AOR] females = 0.53; 95 % Confidence Interval [CI] = 0.45–0.63; AOR males = 0.48; CI = 0.39–0.60), insufficient physical activity, i.e., less than 1 h/day, (AOR females = 0.74; CI = 0.61–0.89; AOR males = 0.73; CI = 0.58–0.92), and insufficient sleep, i.e., less than 7 h/day, (AOR females = 0.56; CI = 0.45–0.68; AOR males = 0.65; CI 0.5–0.85). High family functioning was also associated with lower odds of being overweight/obese (AOR = 0.73; CI = 0.60–0.88) and eating fast food one or more times/week (AOR = 0.74; CI = 0.61–0.89) among females only. Among females, high-quality mother and father relationship were both associated with lower odds of being overweight/obese and disordered eating, eating fast food, and insufficient sleep and the magnitude of associations were similar for mother and father relationship quality (AOR range 0.61–0.84). Among males, high-quality mother and father relationship were both associated with lower odds of disordered eating, insufficient physical activity and insufficient sleep, but only father relationship quality was associated with lower odds of overweight/obesity. Conclusions: Adolescents/young adults reporting high family functioning and more positive relationships with their parents reported better weight-related behaviors. For weight status, females appear to be affected equally by the quality of their relationship with both parents, whereas males may be more affected by their relationship with fathers.

Family functioning and quality of parent-adolescent relationship: cross-sectional associations with adolescent weight-related behaviors and weight status
Adjusting models of ordered multinomial outcomes for nonignorable nonresponse in the occupational employment statistics survey

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

An establishment’s average wage, computed from administrative wage data, has been found to be related to occupational wages. These occupational wages are a primary outcome variable for the Bureau of Labor Statistics Occupational Employment Statistics survey. Motivated by the fact that nonresponse in this survey is associated with average wage even after accounting for other establishment characteristics, we propose a method that uses the administrative data for imputing missing occupational wage values due to nonresponse. This imputation is complicated by the structure of the data. Since occupational wage data is collected in the form of counts of employees in predefined wage ranges for each occupation, weighting approaches to deal with nonresponse do not adequately adjust the estimates for certain domains of estimation. To preserve the current data structure, we propose a method to impute each missing establishment’s wage interval count data as an ordered multinomial random variable using a separate survival model for each occupation. Each model incorporates known auxiliary information for each establishment associated with the distribution of the occupational wage data, including geographic and industry characteristics. This flexible model allows the baseline hazard to vary by occupation while allowing predictors to adjust the probabilities of an employee’s salary falling within the specified ranges. An empirical study and simulation results suggest that the method imputes missing OES wages that are associated with the average wage of the establishment in a way that more closely resembles the observed association.

Adjusting models of ordered multinomial outcomes for nonignorable nonresponse in the occupational employment statistics survey
R markdown: Integrating a reproducible analysis tool into introductory statistics

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Nolan and Temple Lang argue that “the ability to express statistical computations is an es- sential skill.” A key related capacity is the ability to conduct and present data analysis in a way that another person can understand and replicate. The copy-and-paste workflow that is an artifact of antiquated user-interface design makes reproducibility of statistical analysis more difficult, especially as data become increasingly complex and statistical methods become increasingly sophisticated. R Markdown is a new technology that makes creating fully-reproducible statistical analysis simple and painless. It provides a solution suitable not only for cutting edge research, but also for use in an introductory statistics course. We present experiential and statistical evidence that R Markdown can be used effectively in introductory statistics courses, and discuss its role in the rapidly-changing world of statistical computation.

R markdown: Integrating a reproducible analysis tool into introductory statistics
Analysis of partially observed clustered data using generalized estimating equations and multiple imputation

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Clustered data arise in many settings, particularly within the social and biomedical sciences. For example, multiple-source reports are commonly collected in child and adolescent psychiatric epidemiologic studies where researchers use various informants (for instance, parents and adolescents) to provide a holistic view of a subject’s symptoms. Fitzmaurice et al. (1995, American Journal of Epidemiology 142: 1194–1203) have described estimation of multiple-source models using a standard generalized estimating equation (GEE) framework. However, these studies often have missing data because additional stages of consent and assent are required. The usual GEE is unbiased when data are missing completely at random in the context of Little and Rubin (2002, Statistical Analysis with Missing Data [Wiley]). This is a strong assumption that may not be tenable. Other options, such as the weighted GEE, are computationally challenging when missingness is nonmonotone. Multiple imputation is an attractive method to fit incomplete data models while requiring only the less restrictive missing-at-random assumption. Previously, estimation of partially observed clustered data was computationally challenging. However, recent developments in Stata have facilitated using them in practice. We demonstrate how to use multiple imputation in conjunction with a GEE to investigate the prevalence of eating disorder symptoms in adolescents as reported by parents and adolescents and to determine the factors associated with concordance and prevalence. The methods are motivated by the Avon Longitudinal Study of Parents and their Children, a cohort study that enrolled more than 14,000 pregnant mothers in 1991–92 and has followed the health and development of their children at regular intervals. While point estimates for the missing-at-random model were fairly similar to those for the GEE under missing completely at random, the missing-at-random model had smaller standard errors and required less stringent assumptions regarding missingness.

Analysis of partially observed clustered data using generalized estimating equations and multiple imputation
Handling missing data in RCTs: A review of the top medical journals

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Background--Missing outcome data is a threat to the validity of treatment effect estimates in randomized controlled trials. We aimed to evaluate the extent, handling, and sensitivity analysis of missing data and intention-to-treat (ITT) analysis of randomized controlled trials (RCTs) in top tier medical journals, and compare our findings with previous reviews related to missing data and ITT in RCTs. Methods--Review of RCTs published between July and December 2013 in the BMJ, JAMA, Lancet, and New England Journal of Medicine, excluding cluster randomized trials and trials whose primary outcome was survival. Results--Of the 77 identified eligible articles, 73 (95%) reported some missing outcome data. The median percentage of participants with a missing outcome was 9% (range 0 – 70%). The most commonly used method to handle missing data in the primary analysis was complete case analysis (33, 45%), while 20 (27%) performed simple imputation, 15 (19%) used model based methods, and 6 (8%) used multiple imputation. 27 (35%) trials with missing data reported a sensitivity analysis. However, most did not alter the assumptions of missing data from the primary analysis. Reports of ITT or modified ITT were found in 52 (85%) trials, with 21 (40%) of them including all randomized participants. A comparison to a review of trials reported in 2001 showed that missing data rates and approaches are similar, but the use of the term ITT has increased, as has the report of sensitivity analysis. Conclusions--Missing outcome data continues to be a common problem in RCTs. Definitions of the ITT approach remain inconsistent across trials. A large gap is apparent between statistical methods research related to missing data and use of these methods in application settings, including RCTs in top medical journals.

Handling missing data in RCTs: A review of the top medical journals
Prevalence of purging at age 16 and associations with negative outcomes among girls in three community-based cohorts

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Background--The comorbidity of purging behaviours, such as vomiting, inappropriate use of laxatives, diuretics or slimming medications, has been examined in literature. However, most studies do not include adolescents, individuals who purge in the absence of binge eating, or those purging at subclinical frequency. This study examines the prevalence of purging among 16-year-old girls across three countries and their association with substance use and psychological comorbidity. Methods--Data were obtained by questionnaire in 3 population-based cohorts (Avon Longitudinal Study of Parents and Children (ALSPAC), United Kingdom, n = 1,608; Growing Up Today Study (GUTS), USA, n = 3,504; North Finland Birth Cohort (NFBC85/86), Finland, n = 2,306). Multivariate logistic regressions were employed to estimate associations between purging and outcomes. Four models were fit adjusting for binge eating and potential confounders of these associations. Results--In ALSPAC, 9.7% of girls reported purging in the 12-months prior to assessment, 7.3% in GUTS, and 3.5% in NFBC. In all 3 cohorts, purging was associated with adverse outcomes such as binge drinking (ALSPAC: odds ratio (OR) = 2.0, 95% confidence interval (CI) = 1.4–2.9; GUTS: OR = 2.5, 95% CI = 1.5–4.0; NFBC: OR = 1.7, 95% CI = 1.0–2.8), drug use (ALSPAC: OR = 2.9, 95% CI = 1.8–4.7; GUTS: OR = 4.5, 95% CI = 2.8–7.3; NFBC: OR = 4.1, 95% CI = 2.6–6.6), depressive symptoms in ALSPAC (OR = 2.2, 95% CI = 1.5–3.1) and GUTS(OR = 3.7, 95% CI = 2.2–6.3), and several psychopathology measures including clinical anxiety/depression in NFBC (OR = 11.2, 95% CI = 3.9, 31.7). Conclusion--Results show a higher prevalence of purging behaviours among girls in the United Kingdom compared to those in the United States and Finland. Our findings support evidence highlighting that purging in adolescence is associated with negative outcomes, independent of its frequency and binge eating.

Prevalence of purging at age 16 and associations with negative outcomes among girls in three community-based cohorts
A latent class analysis to empirically describe eating disorders through developmental stages

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Objectives--The current standards for classifying eating disorders were primarily informed by adult, clinical study populations, while it is unknown whether an empirically based classification system can be supported across preadolescence through young adulthood. Using latent class analyses, we sought to empirically classify disordered eating in females from preadolescence to young adulthood, and assess the association between classes and adverse outcomes. Method--Latent class models were fit using observations from the 9,039 girls participating in the growing up today study, an on-going cohort following participants annually or biennially since 1996 when they were ages 9–14 years. Associations between classes and drug use, binge drinking, and depressive symptoms were assessed using generalized estimating equations. Results--Across age groups, there was evidence of six classes: a large asymptomatic class, a class characterized by shape/weight concerns, a class characterized by overeating without loss of control, and three resembling full and subthreshold binge eating disorder, purging disorder, and bulimia nervosa. Relative prevalences of classes varied across developmental stages, with symptomatic classes increasing in prevalence with increasing age. Symptomatic classes were associated with concurrent and incident drug use, binge drinking, and high depressive symptoms. Discussion--A classification system resembling broader definitions of DSM-5 diagnoses along with two further subclinical symptomatic classes may be a useful framework for studying disordered eating among adolescent and young adult females.

A latent class analysis to empirically describe eating disorders through developmental stages
Web-based alcohol screening and brief intervention for university students: A randomized trial

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Importance--Unhealthy alcohol use is a leading contributor to the global burden of disease, particularly among young people. Systematic reviews suggest efficacy of web-based alcohol screening and brief intervention and call for effectiveness trials in settings where it could be sustainably delivered. Objective--To evaluate a national web-based alcohol screening and brief intervention program. Main Outcomes and Measures--A fully automated 5-month follow-up assessment was conducted that measured 6 primary outcomes: consumption per typical occasion, drinking frequency, volume of alcohol consumed, an academic problems score, and whether participants exceeded medical guidelines for acute harm (binge drinking) and chronic harm (heavy drinking). A Bonferroni-corrected significance threshold of .0083 was used to account for the 6 comparisons and a sensitivity analysis was used to assess possible attrition bias. Conclusions and Relevance--A national web-based alcohol screening and brief intervention program produced no significant reductions in the frequency or overall volume of drinking or academic problems. There remains a possibility of a small reduction in the amount of alcohol consumed per typical drinking occasion.

Web-based alcohol screening and brief intervention for university students: A randomized trial
Assessing eating disorder symptoms in adolescence: Is there a role for multiple informants?

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Objectives--Epidemiologic studies of adolescent psychiatric disorders often collect information from adolescents and parents, yet most eating disorder epidemiologic studies rely only on adolescent report. Methods--We studied the eating disorder symptom reports provided by 7,968 adolescents from the Avon Longitudinal Study of Parents and Children (ALSPAC), and their parents, who were sent questionnaires at participants’ ages 14 and 16 years. Both adolescents and parents were asked questions about the adolescent's eating disorder symptoms, including binge eating, vomiting, laxative use, fasting, and thinness. We assessed the concordance of parent and adolescent report cross-sectionally using kappa coefficients, and further looked at how the symptom reports were predictive of adolescent body mass and composition measured at a clinical assessment at 17.5 years. Generalized estimating equations were used to model the symptom reports’ associations with risk factors and clinical outcomes. Results--Parents and adolescents were largely discordant on symptom reports cross-sectionally (kappas0.3), with the parent generally less likely to report bulimic symptoms than the adolescent but more likely to report thinness. Female adolescents were more likely to report bulimic symptoms than males (e.g., 2-4 times more likely to report binge eating), while prevalence estimates according to parent reports of female vs. male adolescents were similar. Both parent and adolescent symptom reports at ages 14 and 16 years were predictive of age-17.5 body mass and composition measures; parentally-reported binge eating was more strongly predictive of higher body mass and composition.

Assessing eating disorder symptoms in adolescence: Is there a role for multiple informants?
Differential dropout and bias in randomised controlled trials: when it matters and when it may not

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Dropout in randomised controlled trials is common and threatens the validity of results, as completers may differ from people who drop out. Differing dropout rates between treatment arms is sometimes called differential dropout or attrition. Although differential dropout can bias results, it does not always do so. Similarly, equal dropout may or may not lead to biased results. Depending on the type of missingness and the analysis used, one can get a biased estimate of the treatment effect with equal dropout rates and an unbiased estimate with unequal dropout rates. We reinforce this point with data from a randomised controlled trial in patients with renal cancer and a simulation study.

Differential dropout and bias in randomised controlled trials: when it matters and when it may not
High shape concerns predicts becoming obese, binge drinking, and drug use among adolescent and young adult males

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

To investigate whether males with psychiatric symptoms related to disordered eating and concern about physique are more likely to become obese, to start using drugs, to consume alcohol frequently (binge drinking), or to develop high levels of depressive symptoms.

High shape concerns predicts becoming obese, binge drinking, and drug use among adolescent and young adult males
Setting the stage for data science: Integration of data management skills in introductory and second courses in statistics

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

Many have argued that statistics students need additional facility to express statistical computations. By introducing students to commonplace tools for data management, visualization, and reproducible analysis in data science, and applying these to real-world scenarios, we prepare them to think statistically. In an era of increasingly big data, it is imperative that students develop data-related capacities, beginning with the introductory course. We believe that the integration of these precursors to data science into our curricula—early and often—will help statisticians be part of the dialogue regarding Big Data and Big Questions.

Setting the stage for data science: Integration of data management skills in introductory and second courses in statistics
Data science in statistics curricula: Preparing students to "think with data"

Horton, Nicholas J. (Department of Mathematics and Statistics, Amherst College)

A growing number of students are completing undergraduate degrees in statistics and entering the work force as data analysts. In these positions, they are expected to understand how to utilize databases and other data warehouses, scrape data from Internet sources, program solutions to complex problems in multiple languages, and think algorithmically as well as statistically. These data science topics have not traditionally been a major component of undergraduate programs in statistics. Consequently, a curricular shift is needed to address additional learning outcomes. The goal of this paper is to motivate the importance of data science proficiency and to provide examples and resources for instructors to implement data science in their own statistics curricula. We provide case studies from seven institutions. These varied approaches to teaching data science demonstrate curricular innovations to address new needs. Also included here are examples of assignments designed for courses that foster engagement of undergraduates with data and data science.

Data science in statistics curricula: Preparing students to "think with data"