Using EndNote

This summer I'm teaching a class on research and writing that is used to prepare our IT Leadership graduate students for their Master's thesis.  In this class we review the research skills they have developed in their classes and advance these skills so they will be ready for their thesis.  We also work on enhancing their scholarly writing to stand the more rigorous requirements of the published thesis.

This year I have encouraged our students to  enlist the use of bibliographic software, like EndNote, to help organize their literature sources and ease the burden for formatting all of their references.  I use EndNote consistently with my research and find it valuable to refer back to articles I have read and to store my notes for these articles.  I hope to instill similar patters for my students.

One thing that seems to get in the way for my students application of this software is the automated cite-while-you-write functions EndNote has with Word.  This seems to be adding complexity on using the software that scares students away.  However, if we use a more manual integration of EndNote with our writing the software is very easy to use.  In order to encourage this more simplistic approach to using this bibliographic software I created a video demonstrating how I configure and use EndNote for my research.

Hopefully, a more basic application of the bibliographic software will encourage students to apply the software to help them manage their literature sources and enable them to be more efficient in building a literature review for their research papers.

Creating a Database

Earlier this month I posted about the Dodgers traffic patterns data source I am using from the University of California Irvine for my database course.  This data source provides over 50,000 rows of data I can use for SQL exercises in my course.  Now that I have the data files, I need to load this into a relational database.  This is where the fun begins.

In the past, I created database scripts so that we are able "reset" the database to its original state.  This reset works well for when students begin adding, deleting, or modifying records.  I typically write a database script to drop each of the database tables, create new tables, and insert records into the database.  In the past, when inserting records, I created up to two dozen insert statements to populate the tables.  However, with this much larger database I did not want to write the 50,000 insert statements.  I could use a data loading tool but I want to be able to reset the database and allow students to create this same database on their own for practice.  This created a challenge.

I still needed to create the 50,000 insert statements but rather than writing these statements myself, I enlisted the help of Excel.  Opening the individual data files from the data source text file into Excel, I created formulas to concatenate pieces of the SQL insert statement together with the data to create a complete insert statement for the first record in the first row of the spreadsheet.  I then copied this same concatenation formula across all 50,000 rows and I ended up with a SQL insert statement for each row.  I then copied the set of insert statements to my script and ran it in database environment (after testing a single generated insert statement).  No errors, and all records were created.

While most of the time was spent preparing the data for this automated SQL statement generator (ensuring the date field stayed formatted as a date was a challenge), I was able to create the database script within a couple hours.  I now have a database script that will drop the existing tables, create new tables, and insert all 50,000 rows into the tables.  The script is about 4.5MB (a lot of SQL commands) and takes approximately 10 seconds to execute on my laptop.  I'm not ready to begin using this script with my students.

Knowledge Management in Education

My main scholarly interest lies in the field of knowledge management.  Particularly in knowledge management systems and the structuring of explicit knowledge.  Yesterday I read an article about the adoption of corporate knowledge management practices to higher education.  The article was written back in 2000, so a lot has changed since then but I was also amazed by how little we have advanced in the knowledge management field.

The big change I noticed has been the emergence of a popular knowledge management application to support both tacit and explicit knowledge.  We now have software to help us support knowledge management but there is still a long ways to go for this field to mature and the software to properly facilitate the process.

While knowledge is central to higher education, it is surprising to see how poorly we practice any type of knowledge management activities to develop the faculty and academic programs within the institution.  The knowledge management field appears to be fixated on the corporate setting when there are tremendous opportunities for growth in the academic side.  Perhaps this is what we call a gap in the literature; we need more research done understanding the academic environment in the knowledge management context and applying knowledge management practices to the academic environment.

Team Projects

In my experiences, team projects are a love/hate relationship for students.  Students view these team projects as something different, a way to socialize with classmates, and potentially less work than individual projects.  However, students are often not satisfied with the execution of the project.  Project work is not always distributed evenly across the project team and team members can perform at different levels and have varying degrees of motivation or definitions of success.  These experiences are similar for both undergraduate and graduate students.

While it would be easy to just give up and not use team projects in my classes, there is value in helping the students become more effective working in teams.  While the project teams may not always be successful in the classroom, their lessons from the team project experiences will help the students prepare for working in project teams in the workplace.  For this reason, I plan to continue to use team projects.  However, proper assessment must be adopted in order to increase the student learning from the team experiences and to apply equitable scoring for contributions to the team effort.

I discussed my dilemma with a colleague in the management department earlier this week and we developed an approach to help students reflect on their contributions to the project and to offer a score more indicative of their contribution to the team's deliverables.  In this approach, the teams will earn the same score for the project deliverables but receive separate scores for their contributions to the project.  The score for the individual contributions will be established by the team itself using criteria determined by and agreed upon by the entire class.  The team will be given a total number of points they are to distribute for themselves and they discuss and provide rational why each member should receive the determined point allotment.  The allotted points are added to the deliverable score for each student's project score.

I plan to try this approach to team projects in my undergraduate database modeling course in the Fall semester.  I expect it may require further refinements but I hope to find some level success with this new method of evaluating team projects.

Research and Systems Thinking

I came across an article today that demonstrated a systems thinking approach to problem solving and providing research to help determine the impacts of potential action.  In this case, the researchers studied a downstream effect of Internet privacy restrictions on online advertising.  Rather than simply studying the information available to advertisers, the researchers inquired about the consequences of such restrictions.

The results of the study showed a decrease of up to 65% in effectiveness for the online advertisements.  The common banner advertisements were more affected by privacy restrictions than the larger, and more obtrusive, advertisements.  Looking at this from a systems thinking perspective, restrictions in the consumer data made available to web advertisers resulted in less affective advertisements and will lead to a push to larger and more obtrusive advertisements in order to apply affective online marketing campaigns.  What this means is attempts to limit the data tracked and available to online advertisers could lead to more obtrusive advertisements thus negatively impacting the user experience on the website.

We must then decide if the privacy gained through the restriction of the data is worth the negative user experience.  It may be a good trade-off but we must first understand all of the implications of this action.  We must understand the system to hypothesize the results of any change to the system.  This systems thinking approach will result in better long-term decisions.

Found Some Data

I'm beginning to prepare for a new iteration of my data modeling class for the Fall semester.  I wanted to use a larger dataset than in the past to provide more meaningful queries for class assignments and exercises.  Today I made a step in the right direction.

I came across the Machine Learning Repository site of the University of California - Irvine.  This site houses several examples of datasets that can be used for statistical analysis or other data needs.  One of the datasets I found useful is the Dodger Loop Sensor dataset.  This data comes in two files.  One file contains the game information for each home game during the 2005 baseball season and the second file contains a count of the number of automobiles passing through an exit near Dodger Stadium.  I was able to break-up the game information into game and team tables and then use a third table to hold the traffic information.  Using these three tables, my students will be able to explore the data to find traffic patterns and associate them with the baseball games.

While this is far from perfect, 81 game records and 50,400 traffic records greatly exceed the small (18 records) hotel reservation database I created last year.  I think I will still use both databases but the new database will allow the students to explore the data more than the smaller database.

IT Value

I read an article this week about the end of IT value as a metric for the IT organization.  When I first noticed the article title I was shocked since value is an important measure for the contributions of IT to the organization and the measure indicates alignment with the business strategies of the organization.  However, as the article noted, value is dependent upon what is measured.  In this case, what is measured has become the source of this issue.

Apparently, many organizations are still measuring value in terms of a metric specific to the technology or service itself.  For instance, the help desk may measure value in terms of resolution time or hold time.  However, this is not really value for the organization.  True value should look at satisfaction measure for the end users or the decline in reported incidences.  These true value measures should indicate some type of cost reduction or revenue increase to the organization rather than simply reflecting on an attribute of the service itself.  Value must measure impact to the organization.

Over the past two decades IT has recognized the need to become partners with the functional units of the organization.  We learned we cannot work independent of the organization and we must view IT as a business enabler rather than the creator of elegant technology looking for application.  If we are truly connected to the functional side of the organization, we must measure IT contributions in terms of the impacts to the overall performance of the organization.  Impact to the overall organizational performance represents true IT value.