On normal and justice.
DAT Assignment 2: Chi Square
(For those of you looking for something more interesting, go check out the pictures I took walking around my neighborhood after the blizzard in DC this weekend.)
Following completion of the steps described above, create a blog entry where you submit syntax used to run a Chi-Square Test (copied and pasted from your program) along with corresponding output and a few sentences of interpretation.
The rest is past the jump.
Perspective
I was born and raised and spent a good chunk of my adult life living in the upper Midwest, where it snows regularly and not-infrequently in large quantities. Naturally, I have a number of friends and family who still live there, and unfortunately, apparently also naturally a fair number of them seem to think there’s something they have to feel smugly superior about in watching the mid-Atlantic react to this weekend’s snow. So.. as someone who’s lived in both places, allow me to clarify why that behavior is not only annoying, but also fundamentally flawed and illogical.
If you live somewhere where it snows frequently, it makes sense to build your infrastructure in ways that will withstand sometimes large volumes of snow. This means your city planners were likely intentional about building wider streets and highway shoulders – so that when it snows and the plows come through, there’s room for the resulting snow to be piled up without impinging on traffic. Similarly, your power lines are more likely to be underground, where accumulated ice and heavy snow would be less likely to cause outages. It also makes sense for your cities, counties, and states to invest in sufficient equipment to safely and quickly respond to snow – adequate vehicles to pre-treat all major and most minor roads, enough plows so that even during blizzards they can make passes through most streets multiple times. Because if you live somewhere where it snows frequently, *not* doing those things would be irresponsible, both in terms of health and human safety, but also in terms of economic impact on the community.
If you live somewhere where it does not snow frequently, somewhere where in the last three and a half years (as long as I’ve lived in DC) it’s snowed *at all* less than a dozen times and snowed a couple inches at once only one or two times, it does not make sense to build your infrastructure with snow removal in mind. It would be fiscally irresponsible, in fact, for cities, counties, and states to invest in large volumes of equipment for the once-every-half-decade snow storm where they might be justified. It would be more expensive – taking money out of public budgets that could and should be spent on other things more likely to have direct impact on health and safety and economic prosperity – to insist that all power lines be underground, or that all streets be wide enough to accommodate snow banks.
Therefore, when that once-every-half-decade snow storm happens, yes, those places where snow is not a regular occurrence will not be able to respond as efficiently as those places that know and regularly expect snow in large quantities every year. There’s nothing inferior about the preparation or governance of those places that do not regularly experience snow. The people who live there are not inherently stupid, though they may be ignorant of how to drive in so much snow or how to ensure that their home is prepared for what may be several days without the ability to restock *because* it’s not a normal occurrence. Their governments are not overreacting when they close things down for several days to give adequate time and less-trafficked roads to the crews who are working diligently to respond to an abnormal weather event; they’re being responsible and working to keep everyone safe.
Institutional Locus of Control
Those of you in the higher ed world will likely have heard of Florida governor Rick Scott’s invitation to the state’s colleges and universities to a meeting tomorrow with the intent to “challenge” them to 100% post-collegiate employment rates for their Psychology majors. (If you haven’t, Inside Higher Ed wrote it up today.) This prompted some lamenting in higher ed circles, which is not surprising. My thoughts (many of which were posted originally in response to an email about the article linked above on the ASSESS listserv) are below.
Rather than focus on the historical roots and purpose of higher education, I have to wonder if there’s not more of a role our business partners can play (or we could ask them to play) in helping change legislator perspectives about what they need for today’s jobs. It seems that there’s an unhealthy focus on major field of study as the be-all-end-all for determining a graduate’s career success or failure, when numerous surveys of employers tell us that’s not actually what employers care about. For example:
- AAC&U released their latest update to their employer survey about a year ago and found that “[n]early all employers (91%) agree that for career success, ‘a candidate’s demonstrated capacity to think critically, communicate clearly, and solve complex problems is more important than his or her undergraduate major.’” (Emphasis in original.) (This finding was quoted again today, in fact, in an op-ed in Maine that crossed my Twitter feed just now.)
- A similar finding was part of a recent (arguably – 2012) Chronicle of Higher Education employer survey (PDF): “Employers place more weight on experience, particularly internships and employment during school vs. academic credentials including GPA and college major when evaluating a recent graduate for employment.”
There are coalitions of business leaders becoming more active in advocating for better state support of higher education – there’s at least one in Florida called The Florida Council of 100 – though I’m not sure of the agenda of any of them and whether they’re intentional acting on behalf of institutions or higher education as a whole or not. Still, it seems like rather than higher education continuing to beat the same drum, it may be time for a change of tactics and partners to more directly respond to policymakers’ arguably well-meaning but ultimately ill-informed demands.
In addition to enlisting new partners, the academy as a whole could work to raise the visibility of work intended to broaden the focus of higher education outcomes beyond just employment and wages. I’m most familiar with the Post-Collegiate Outcomes (PCO) Initiative completed by the American Association of Community Colleges (AACC), the American Association of State Colleges and Universities (AASCU), and (my employer*) the Association of Public and Land-grant Universities (APLU) last year, which resulted in the PCO Framework and Toolkit designed to help institutions intentionally broaden the conversation about outcomes with their stakeholders. (Full disclosure: I was integrally involved in the PCO Initiative.) The Lumina Foundation recently (and without a lot of fanfare, unfortunately) released It’s Not Just the Money: The Benefits of College Education to Individuals and to Society, which uses data from national sources to document both the financial and human capital benefits to both individuals and their communities from a more highly education population. I also know that there are several institutions who are – as part of their mission – intentionally working to promote the civic engagement outcomes of higher education; James Madison University comes to mind as well as the institutions that are members of The New American Colleges & Universities, which held an event this morning at the National Press Club in DC presumably to engage policy-makers and DC-based think tanks more directly in promoting these outcomes.
There’s still undoubtedly a lot of progress to be made, but I think there are also a lot of potential partners to work with to push back on some of the too-narrowly-focused ideas we’re seeing.
* As always, the opinions expressed in the blog are my own and do not necessarily reflect those my employer.
DMV Assignment 3: Managing Data
Once you have written a successful program that manages your data, create a blog entry where you post your program and the results/output that displays at least 3 of your data managed variables as frequency distributions. Write a few sentences describing these frequency distributions in terms of the values the variables take, how often they take them, the presence of missing data, etc.
More after the jump.. Continue reading
DAT Assignment 1: ANOVA
(For those following along, this is the second course I’m taking to learn SAS. It’s designed to be taken after the first course, but as I’m already familiar with the statistical concepts in both courses, I’m taking them concurrently. That does mean, however, that the order of posts will be somewhat jumpy as I’ll be posting assignments for ANOVA before completing the full set of descriptive analysis required for the first course.)
DMV Assignment 2: My first SAS Program!
Assignment instructions: Following completion of your first program, create a blog entry where you post 1) your program 2) the output that displays three of your variables as frequency tables and 3) a few sentences describing your frequency distributions in terms of the values the variables take, how often they take them, the presence of missing data, etc.
(Apologies for those of you not interested in reading raw SAS code, but for those of you who are, I’m open to suggestions for how to improve. This is fairly basic code, but useful for someone who’s never used SAS.)
Behind the jump for all the details.. Continue reading
DMV Assignment 1: Exploring institutionalized racism: Race and perception of law enforcement and opportunity for achievement between Blacks and Whites during the beginning of the #BlackLivesMatter movement
(A general note: I’m taking this course mostly for the opportunity learn SAS for data analysis. That said, my intent is to keep up with the course assignments, though likely with a lower level of rigor than desired by the instructors. While the topic I’ve chosen is of personal interest, and has some bearing on my professional work in higher education, it’s not within my main professional focus. Therefore, I will be cutting corners on conducting a full and proper lit review and may take the liberty of similar shortcuts in later assignments.)
Of the five data sources made available through the course, I’ve chosen to use the Outlook on Life Surveys data. Conducted in 2012 – in the midst of the recent spike in attention and interest by mainstream media in systemic/institutionalized racism prompted by the creation of the #BlackLivesMatter movement – the Outlook Surveys consist of responses to two internet surveys fielded between August and December 2012. The first survey yielded 2,294 responses (a 55.3% response rate) from a nationally representative panel of US adults (aged 18 and older) divided into four groups: African American/Black males, African American/Black females, White/other race males, and White/other race females. The survey sample contained a large oversample of African American/Black respondents. Panel members (and therefore members of the included sample) were randomly recruited using random-digit dialing and address-based sampling methodologies and households, where necessary, were provided with access to the internet and hardware to do so. The second wave of the survey consisted of interviews with 1,601 of the respondents from the first wave (a 75.1% response rate).
I’m most interested in exploring the relationships between race, social class, respondents perception of opportunity for success, and relations with law enforcement. Given a variety of evidence supporting a pattern of systemic racism (link is to a PDF) in the United States, specifically manifesting in police interactions with African Americans/Blacks, and the increased attention by mainstream media drawn to incidents of unequal response by police to incidents involving African Americans/Blacks starting with the death of Trayvon Martin, I expect to find that relations with law enforcement for African Americans/Blacks will be significantly more negative than for their white counterparts, and that, in particular, African American/Black males will report more negative relations than African American/Black females. I also expect to find interactions that may mitigate or exacerbate relations with law enforcement related to social class. Specifically, I expect that respondents who are both African American/Black and of lower social class will also report more negative relations with law enforcement. Further, given that systemic racism is not limited to interactions with law enforcement, but rather embedded and ingrained in our culture more broadly, I expect to find that African American/Black respondents also expressed lower expectations about their opportunities for success.
There are a number of items included in the Outlook Surveys that relate to the variables I am most interested in. By way of example, an indicator of social class could be created by combining responses to questions on personal income sources, access to cable television and the internet, reliance on social services, highest level of education received, and/or home ownership status. Similarly, perception of opportunities for success could include respondent’s beliefs regarding their personal current and/or expected progress toward achieving the American dream, their expectations of success for their children or their children’s children, and/or beliefs about the relative equality of opportunity for Blacks and Whites. Relations with law enforcement are (fortunately) covered with only a small number of questions on the survey, though this simplicity may ultimately result in too broad a brush to describe the relationships in which I’m interested with sufficient specificity.
As I begin to work more directly with the data from the Outlook Surveys in the coming weeks, I expect that my variable selection will become more refined. For now, I have reviewed the codebook for the Outlook Surveys and highlighted those variables most likely to be of interest as I proceed. I expect that additional exploration of the data, specifically disaggregating the responses by race and gender into the four groups included in the responses, will prompt additional questions and the need for further review of the available data.
Fair warning
For various and sundry reasons, I’m participating in a few MOOCs this month on data analysis and interpretation. Yes, I already know these things and so the MOOCs are at best refresher content. However, we are likely becoming a SAS shop at work, which I’ve never had to learn, and the MOOCs – offered through Coursera by Wesleyan* – include specific guidance on learning SAS for data analysis.
One of the requirements of the courses is to maintain a public blog for various assignments related to the courses. I though briefly about starting a new tumblr for this purpose, but decided that adding another social media outlet to my existing bouquet might just put me over the edge. Therefore, I’ll be posting assignments here. You’re all welcome to comment (either directly on the blog or on the posts that push to Facebook or Twitter) if you wish, or to ignore them if you wish (I should hope that’s fairly obvious generally).
I’m not sure how frequently I’ll need to post for the courses, or whether I’ll work in posts between the required ones on other topics throughout the next month. The next post, though, will include information on the research question and data set I’ll be using for the courses, so you’ll know more then about whether you’ll want to tune in or tune out in the interim.
* Wesleyan offers the courses free if you take them as individual courses**, or you can choose to pay a certification fee to earn a specialization in Data Analysis and Interpretation. At the moment, I’m opting not to pay the fee for the specialization – it’s hopefully duplicative given my Masters degree! – but you, of course, have the option to “upgrade” later.
** The courses I’m taking (all at once because why not?) are: