The percent of graduates who complete a degree in four years from start to finish is not the same thing as a four-year graduation rate.

The former measure is a backward-looking measure. It looks at all the graduates in a given year and then checks back to see what percentage started at the institution within the previous four years.

The latter measure, a four-year graduation rate, is a forward-looking measure in that it starts with a group or cohort of similar students (in this case, first-time in college, full-time at first enrollment) and checks to see what percentage of these students completed their degree within four years.

Why does this matter?

Four years ago,  a Small Liberal Arts College (a private, nonprofit SLAC) announced a new commitment to (certain) students: The college announced this week it will waive tuition costs for any additional courses needed to complete a degree if a student isn’t able to graduate in four years, provided certain requirements are met. The article goes on to report, based on the announcement on the college’s website, that 95% of graduates complete their degree within four years and that this is a much higher rate than the four-year completion rate for all graduates at private colleges (almost 80%) and for public institutions (below 50%). The source of the data is the National Association of Independent Colleges and Universities (NAICU).

In fact, one of the financial aid pages on the College’s website goes so far as to say this:

Many consider state colleges as having a low sticker price. In reality, once we factor in aid, the final cost isn’t significantly different than a state college. It is also important to note that 90% of our students graduate in four years saving the cost of a 5th year.

The College and NAICU should know better. The comparisons they are making are not appropriate. They are, in fact, deceitful. Note: We should probably be fair to NAICU as we do not know they were complicit in this deceit.

So, what is the real story here?

First, we assume that nearly all students graduating from the College  probably do graduate within four-years. It is simply too expensive not do so.

The National Center for Education Statistics (NCES) maintains an excellent website,College Navigator, that is quite useful for finding and comparing colleges. It’s data is based on annual surveys, the Integrated Postsecondary Education Data System (IPEDS), required of all Title IV (federal student aid programs) participating institutions. It reveals that only 49% of all the students at the College starting in the Fall of 2008 graduated in four-years or less. Only 62% graduated within six years.

This is a far cry from 95%.
If the statements on the College’s website are representative of the critical thinking and analysis skills taught at the college, we have a serious problem. I kind of hope these are cynical, deceitful statements of an institution trying to stay afloat. That can be dealt with greater ease than an inability to engage in critical thinking because I just don’t think they understand the problem.

The College seems to be intent on putting the slack into SLAC.

Just as a reminder, here is the Principle of Integrity from Southern Association of Colleges and Schools Commission on Colleges:

Integrity, essential to the purpose of higher education, functions as the basic contract defining the relationship between the Commission and each of its member and candidate institutions. It is a relationship in which all parties agree to deal honestly and openly with their constituencies and with one another. Without this commitment, no relationship can exist or be sustained between the Commission and its accredited and candidate institutions.

Integrity in the accreditation process is best understood in the context of peer review, professional judgment by peers of commonly accepted sound academic practice, and the conscientious application of the Principles of Accreditation as mutually agreed upon standards for accreditation. The Commission’s requirements, policies, processes, procedures, and decisions are predicated on integrity.

The Commission on Colleges expects integrity to govern the operation of institutions and for institutions to make reasonable and responsible decisions consistent with the spirit of integrity in all matters. Therefore, evidence of withholding information, providing inaccurate information to the public, failing to provide timely and accurate information to the Commission, or failing to conduct a candid self-assessment of compliance with the Principles of Accreditation and to submit this assessment to the Commission, and other similar practices will be seen as the lack of a full commitment to integrity. The Commission’s policy statement “Integrity and Accuracy in Institutional Representation” gives examples of the application of the principle of integrity in accreditation activities. The policy is not all-encompassing nor does it address all possible situations. (See Commission policy “Integrity and Accuracy in Institutional Representation.”) Failure of an institution to adhere to the integrity principle may result in a loss of accreditation or candidacy.

1.1 The institution operates with integrity in all matters. (Integrity)

(Note: This principle is not addressed by the institution in its Compliance Certification.)

Update to the Embedded Challenge

Turns out Jon’s problem wasn’t nearly that difficult..once I understood it.

Specifically, his challenge is with and the Tableau share links. Once I played around wit both sites and did a quick search to verify my suspicions, it was a piece of cake. allows use of iframes to display content from other servers. While it does allow access to raw HTML, it does not allow inclusion of script tags, so you are left with using the iframe tool and source URL.  When I applied a critical eye to the two share options, I noted:

<script type='text/javascript' src=''></script>
<div class='tableauPlaceholder' style='width: 800px; height: 700px;'>
<noscript><a href=''>
<img alt='Dashboard 3 ' src='' style='border: none' /></a></noscript>




I noted the parameter showVizHome was present in the embed code, but not the URL. A quick search finds this helpful page which allows Jon and anyone else t resolve the problem by adding &:showVizHome=no to the end of URL and problem is solved.

An Embedded Challenge

It started with a tweet:

Okay, this actually seemed pretty simple and straightforward to me, and I said as much. I don’t use Tableau, I use LogiAnalytics, but a javascript embed string is a javascript embed string. A string is a string. Isn’t that the basis of string theory?

Plus, I had already done something similar to build a very light presentation engine using web pages and Logi reports. We had built two different solutions.

The specs are these:

1) A way to use the embed code for LogiAnaylitics or Tableau in a presentation, typically PowerPoint. (My first thought was to use MS Access since it has a web browser object available in the forms toolset.)

2) “Share links” were not an adequate solution because they pull the entire page, not just the report/visualization. (This doesn’t actually apply to Logi. Logi is a BI engine based in .Net and the resulting in XML files are processed through a server application overlaying IIS, and it is fed with a report definition and all requisite parameters processed as a URL querystring. ie )

3) It has to be easy to use, not requiring any scripting – by the user.

A quick web search turns up Live Web which is promising. A little more searching and one finds this is not the first attempt to address the challenge – here is a step-by-step approach that may meet Jon’s needs, except he would need to code it. It also uses the links instead of the embed code, so it probably doesn’t give exactly what he wants. But…why do this? Since I am not a Tableau user, why do I care?

As I said, there are similarities in using Logi and Tableau, and there is something powerful about doing a data-heavy presentation without relying on either static images or exiting to a web browser. This is why I created a web-based presentation model to give presentations off the SCHEV Research web site using a defined series of live pages with predetermined parameters for the default views of each page. We built the Pathways model to focus on delivering sequential Logi reports that needed only parameter changes (handle via global variable settings) to be updated while having the flexibility of using existing report objects or creating new objects that could be used elsewhere.  The slick thing about the embed code is being able to frame around an active BI object and explore the data while in the midst of a presentation without losing the rhythm of progress forward.

I downloaded LiveWeb after my first attempts with MS Access 2013 failed. I kept running into an inability to load the web browser object….let me explain the problem with 2013.

I isolated the project into two parts, storage and display. Any project that comes along, what’s the first thing to be done? Why, build a database! Actually, all that is needed is a table and this could be in Excel. Of course, it could also be in just a text file, but I am database guy. So, I built a table to hold the embed codes and associate them with username and a presentation name, so that multiple users could use this tool for multiple presentations.

I then created a form in Access and added the web browser control to it and it failed. I did some research and found registry settings that I would need to change in Windows to make this work and then tried again with Access  2010 since the web pages  I had found suggested it was a security change in 2013 creating the problem and I wanted to keep the solution as simple as possible. So, I tried 2010 and could install the browser control. Unfortunately, it was limited in that it has to be passed a URL, not just HTML code. The easiest solution would be something that allows passing something like this:

<HTML><BODY><div id="divFA19_Report" data-applicationUrl="" data-report="FA19_Report" data-autoSizing="height" data-linkParams="{'lbUNITID':'XXXALL', 'lbFTEGRP':'1'}" ></DIV></BODY></HTML>


<HTML><BODY><script type='text/javascript' src=''></script><div class='tableauPlaceholder' style='width: 1024px; height: 694px;'><noscript><a href=''><img alt='Dashboard 1 ' src='' style='border: none' /></a></noscript>  </div></BODY></HTML>

Even better would be to lose the need to wrap the code with the HTML and BODY tags.

I created a model that did exactly that. My approach is to coding is to take advantage of all sorts of tools I have available. One of which is that I have multiple web servers under my control and thus it is way easy to create a very simply Classic ASP page that received the embed code as a request variable (in the URL querystring), place it in the page, and execute it. Unfortunately, the Tableau code tends to exceed the limits that a number of browsers will allow for that, servers as well, as a good security measure. So, enter the database.

We have a table with codes. We simply need to access that, pull one row at a time, load it in the page, and execute the page and we are done. I moved the data to a table on SQL Server and created a simple stored procedure.

 [key] [int] IDENTITY(1,1) NOT NULL,
 [usrname] [varchar](50) NULL,
 [presentation_name] [varchar](50) NULL,
 [slide_title] [varchar](50) NULL,
 [slide_order] [int] NULL,
 [embed_code] [varchar](max) NULL,
 [embed_type] [varchar](50) NULL

The stored procedure is, as I said, quite simple:

ALTER proc [dbo].[sp_embed_slideshow] @UserNM as varchar(5), @prestitle as varchar(50), 
@slide as int
Select key, usrname, presentation_name, slide_title, slide_order, embed_code, embed_type 
from slideshow
where usrname=@userNM and presentation_name=@prestitle and slide_order=@slide

The next piece is a simple Classic ASP web page using VBScript:

<!--#INCLUDE VIRTUAL="/someconnection.asp"-->
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "">
<html xmlns="">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<body bgcolor="#ffffff">
 cmdtext="exec sp_embed_slideshow '" & user_name & "','" & presentation_name & "'," & slide_num
 dim rget
 set rget =CreateObject("ADODB.Recordset") cmdtext, someconnection
 do while not rget.eof
response.write (embed_code)
<!--THIS IS FOR LOGIANALYTICS applications only-->
<script src="" type="text/Javascript"> </script>

Once I had downloaded LiveWeb and loaded into a presentation, all I had to was step through it’s wizard and load the following URLS:

These next links use code from Jon’s site:

Clearly, this is not the solution desired. It requires some programming skills (although all the real work is done and anything else is window-dressing or ease-of-use enhancements). It also requires access to a web server and database server.


One can create this in MS Access and then activate IIS (or install it if it is not already installed on your WIndows machine) and use your desktop or laptop for a local user. I don’t really recommend this approach as learning IIS 7 and above is an unnecessary pain for just this project. The alternative is to download UltiDeve Cassini. install it, create a data connection between Cassini and Access, use the ASP code above, and you’re off to the races!

Some night this week I will try to finish off the Access model for download for those that wish to play.

There are probably dozens of other ways to do this, perhaps one or two very easy ways that I missed. Maybe I will create a private service for institutional researchers to use for a very modest fee if there is interest.

I’m interested in any reactions or alternatives.

Will, this is the post I suggested you might want to share.

UPDATE: The needed solution was actually much simpler, once I understood the real problem.

The Beast that Shouted “Notice Me!” at the Heart of the Web

With apologies to Harlan Ellison. While the story “The Beast that Shouted Love at the Heart of the World” itself has not stuck with me over the years, the title certainly has, certainly not the same way that I Have No Mouth, and I Must Scream has done so.  Regardless, the question is this. Is “”Notice Me!” all that much different from Love?

In the 90s, I played with a lot of different technologies on the web. I wrote a fair amount that I never put online nor tried to publish. I actually had more time then than I do now. I don’t  know that I have all that much more to say, but I seem to have a greater desire now to say it or to get document my thoughts.

When we rolled out our revised/renewed website in the fall of 2012, we built social community structure into it with a blog, wiki, and user forum. The forum has not taken off at all, but this did not surprise me. The institutional research community has never been much on such things, and that has been our primary target audience.  I started the blog over there with one thing in mind, but it has morphed into something else as I have blogged there about weekly for most of the year.

At the same time, I created the @SCHEVResearch Twitter account and used it very little for several months. I figured initially we would just push notices that out that way to supplement other avenues. As I spent more time on Twitter, I found out just how useful it is to connect with some really smart people doing good work and stay current on what is going on. Then I began to see the need for a more personal interaction on Twitter that was less appropriate for the account representing a department in a state agency, so I dusted off the unused personal account and announced that I would be moving some activity there.

So, as I eagerly scan for page view accounts and new followers, I wonder two things.

Am I shouting “Notice Me!” loudly enough?

Why, oh why, am I shouting “Notice Me!”?

I think the answer to my first question is probably that I am not. I need to tweet more, follow more people, use more hashtags, all that stuff. Part of the problem is that I still have to be mindful of the limits of free speech for a public official. I censor myself a lotto provide the illusion that I am a serious, thoughtful person.

As to why shout “Notice Me!” that is easy. I am little tired of the lack of awareness of some of things we are doing, that I am doing. For example, I get kind of annoyed when faculty bloggers from Virginia institutions or presidents of Virginia institutions who write op-ed pieces attacking PayScale or USED’s ratings proposal to include earnings measures fail to mention what we are doing in Virginia. While I am glad they are not attacking our work (although some of the presidents seem to like to do that face-to-face) they could at least demonstrate an awareness of what we have done and why it is different. Ignoring it is not going to make it go away – it is the law.

As is the new requirement that all colleges and universities receiving general fund dollars (with direct appropriation or student aid) from the state shall post a link on their websites to our reports (the italics represent new language):

“for each degree awarded by each institution and shall, at a minimum, include the percentage of graduates known to be employed in the Commonwealth, the average salary, and the average higher education-related debt for the graduates on which the data is based; rates of enrollment in remedial coursework for each institution; individual student credit accumulation for each institution; rates of postsecondary degree completion; and any other information that the Council determines is necessary to address adequate preparation for success in postsecondary education and alignment between secondary and postsecondary education.

School divisions and high schools must also provide a link to the reports. Clearly the patron of this bill is paying attention to my work, and more importantly, likes it.

In many ways, the highered world is becoming more socially directed and those that are not paying attention are going to find themselves to be lost. Those that fail to try to be found, will be equally lost. Scott Adams once suggested in either The Dilbert Principle or The Dilbert Future that the need for a personal webpage may become so pervasive that there would be government subsidies for the homepageless.

I’ve taken a hiatus from membership and involvement with Association for Institutional Research. For many years, that was a large part of my professional life. I’m returning to the Forum this year, and we will see how that goes.

Notice Me!






TANSTAAFL – Obligations of Open Source

Earlier today I was part of Twitter discussion Barmak (@BarmakN) and Chuck (@ShorterPearson) about Heartbleed and the obligations of associated with open source applications.

I feel pretty strongly about this topic. We use one open source software application at work, or rather five. It’s called Sueetie and pulls together several open source projects (wiki, blog, media server, forum) to create a single unified environment. It’s pretty slick and the integration and value-added wrap-arounds make it ideal for our purposes. As soon as we committed to using Sueetie, I joined the user community site, both for help with the set-up and to give back by sharing what we learned along the way. And we learned a lot.

None of my staff are .Net or C# developers, so it was hard to contribute much at that level. However, on the SQL side we could contribute a lot, which we did – right up until the founder pulled the plug. For whatever reason, after several years of development, the community user group never took off. There were only about three active users, including us, plus the founder. There were plenty of downloads, but not much in the way of real involvement. We were also one of the very few to actually buy the license and premium source code.

As a product, Sueetie never really took off. Maybe it was too much effort for most users. One had to know a few things to install it and have decent mastery of the least-used skill these days – RFC (Reading for Comprehension). I think though it also had a lot to do with lack of understanding of how open source works – participation, support, funding – and the desire to get something for nothing. The community of active supporters was not large enough to be self-perpetuating and the founder eventually needed to move his Sueetie-time to revenue generation.

It just doesn’t work that way. There Ain’t No Such Thing As A Free Lunch. TANSTAAFL. People need to read more Robert A. Heinlein. One way or another, things have to be paid for, either with cash or sweat or service or barter.

We’ve stuck with Sueetie. We’ve invested too much effort in using it and developing content. We have all the source code and will begin updating it, focusing on the areas we use most. Some pieces we may ultimately disassociate from our application because we don’t use them, others we will continue to enhance or integrate the newest versions.

The Heartbleed bug is an example of what happens when folks forget TANSTAAFL. Big moneymakers that adopted OpenSSL, but did not take seriously their role in being part of the OpenSSL community have a lot to answer for. Just a reminder these enterprises include Facebook, Google, Yahoo, Amazon, and Instagram. I feel confident they could have added value along the way…especially Facebook and Google.

I guess it is time to pay up.

Open source code is great. It is powerful, can be affordable, and you can customize it. If you need a cheap solution, that’s okay, but you can still contribute to its development through development, sharing your modifications, bug reporting, crowd-funding, letting the developers know how you are using it, and helping others get started. The last is the easiest, and perhaps the most powerful as it can help a user community become large enough to continue.

Remember: identity, community, and stability all apply to the open source movement. Especially if you aren’t paying attention.