Using the Microsoft Bot Framework to Create Educational Bots

Bot Building Steps
Photo by Amber Case

         1.Microsoft QnA Maker
         2.Create a list of frequently asked questions
         4.Student review

Building on the achievements of the innovative chatbots created at Georgia Tech and Georgia State University, the faculty and staff at Brock University’s Centre for Adult Education sought to use the Microsoft Bot Framework to create a bot that would support student success in an online undergraduate degree program.
With the goal of providing students with a means to be able to quickly and efficiently get answers to the questions they may have at any time during the night or day, the program began to develop a list of frequently asked questions. The experiment at Georgia Tech noted that many of the questions that students ask are routine and are usually related to readily available information (Korn, 2016). With this in mind, the student success bot created at Brock University simply provided answers to common questions and included a link to the webpage where the answer was located. This enabled students to easily and quickly find the answer to their question without having to click through numerous pages of the university’s often, difficult to navigate website. Being able to make anonymous inquiries also allowed students to save face and avoid feeling trivial.  Students that used the Georgia State bot noted that they didn’t feel judged for asking what might seem like a “stupid” question. They also appreciated the instantaneous responses, especially when they asked questions at all hours of the night (Herndon, 2017).
Instead of having one person create the bot, a collaborative approach was used to develop the knowledge base that powered the Brock chatbot. This inclusive approach provided the faculty and staff with the opportunity to discuss potential student questions and then consider the answers, as well as to create proactive solutions to common student problems. This resulted in a collaborative professional development opportunity while also serving to develop a comprehensive listing of student services and advising resources that could be used to support individual staff or faculty knowledge development.

The use of Microsoft Office 365 facilitated a collaborative and dynamic process by allowing all contributors to provide their insight into the content and context of the types of questions that were to be included in the knowledge base. The knowledge base was created using an Office 365 Word file that allowed all faculty and staff members to contribute to and edit the content. Sharing access to the bot knowledge base, question log and analytics also serves foster a collaborative approach to the maintenance and review of the bot data. The ability for multiple people to access this data can foster further discussions regarding the when and why students are asking specific questions. 


Further Reading:

        Microsoft Bot Framework
        Microsoft QnA Maker

Flickr Photo by Amber Case

Resources for Creating Successful Graduate Student Reference Letters

One of the essential functions for faculty members is to support the future success of their students. For graduate students, this can entail writing several scholarship reference letters. As these letters can have a significant impact on their ability to engage in high-quality research it is paramount that faculty carefully consider what should be included in these types of letters. Below are links to a number of helpful resources to support the writing excellent reference letters.

Fixing Dangerous Academic Intersections

The installation of a new stop sign at a campus crosswalk provided me the opportunity to reflect on what measures we should take to caution our learners as they traverse through our academic programs. Across any degree program, there can be several dangerous academic intersections that students have to move through as they matriculate. This may be a particularly difficult course or assignment that results in an inordinate number of failures, withdrawals, sick notes or counselling requests.

The challenge is how are we keeping track of these minor and major ‘academic accidents’ and how quickly can we react to mitigate the danger. While I would never advocate for simply removing academic challenges, it is important to reflect on whether we are requiring students to travel down a dark alley, where they are blind to the oncoming dangers, we have created, or should we install a stop sign that requires them to carefully look both ways and thoughtfully consider when and how to proceed.

Each time I approach these stops signs I now find myself reflecting on how can we determine where and when a stops sign may be needed.

What is Your Vision for the Future of Education?

Inspired by Jane McGonigal's SXSWedu "How to Think Like a Futurist" keynote.

  1. Collect signals from the present
  2. Combine signals into a 10-year forecast
  3.  Create personal foresight

Collect Signals

3D printed food +  Virtual Reality + Animal Empathy + Low Carbon Diets =

Combine Signals into a 10-year Forecast

Create Personal Foresight

Future Forecast: Learning is Earning

What is your personal foresight  for the future of education?   What are the positive and negative implications? 

Survey Research and Acceptable Response Rates

The adequacy of response rates to online and paper surveys

Nulty, D. D. (2008). The adequacy of response rates to online and paper surveys: what can be done?. Assessment & Evaluation in Higher Education,33(3), 301-314.

This article is about differences between, and the adequacy of, response rates to online and paper-based course and teaching evaluation surveys. Its aim is to provide practical guidance on these matters. The first part of the article gives an overview of online surveying in general, a review of data relating to survey response rates and practical advice to help boost response rates. The second part of the article discusses when a response rate may be considered large enough for the survey data to provide adequate evidence for accountability and improvement purposes. The article ends with suggestions for improving the effectiveness of evaluation strategy. These suggestions are: to seek to obtain the highest response rates possible to all surveys; to take account of probable effects of survey design and methods on the feedback obtained when interpreting that feedback; and to enhance this action by making use of data derived from multiple methods of gathering feedback. 

When the more traditional and conservative conditions are set, the best reported response rate obtained for on-paper surveys (65%) is only adequate when the class size exceeds approximately 500 students. The best reported response rates for online surveys (47%) are only adequate for class sizes above 750 students. The 20% response rate achieved for online surveys by Griffith University would not be adequate even with class sizes of 2000 students.


Typical Response Rates for Common Survey Types
Surveys that you distribute internally (i.e. to employees) generally have a much higher response rate than those distributed to external audiences (i.e. customers).

Internal surveys will generally receive a 30-40% response rate (or more) on average, compared to an average 10-15% response rate for external surveys.

3 Ways to Improve Your Survey Response Rates

To help improve your survey response rate keep these key factors in mind:

1. Survey Design: Research has shown that surveys should take 5 minutes or less to complete. Although 6 – 10 minutes is acceptable, those that take longer than 11 minutes will likely result in lower response rates. On average, respondents can complete 5 closed-ended questions per minute and 2 open-ended questions per minute.

2. Provide Clear Value: Offer a copy of the final results to all those who complete the survey, and, if appropriate, consider offering an incentive. If you plan to take action based on the results of your survey, make those clear in your survey invitation. Remember, people will be more likely to respond if they understand how that time will be spent.

3. Send Reminders: As the close of your survey approaches, gently nudge those who haven’t finished yet. Limit yourself to no more than two reminder emails, changing the time of day and the day of the week that you send them out so that you can reach as many different respondents as possible.


Survey Sample Size Calculator

When it comes to probability surveying, creating a sample size should never be left to guessing or estimates. Instead, it should be based on three criteria:

The size of your target population: This refers to the total amount of people that are eligible to participate in your survey. For example, a study on Ontario citizens’ sleeping habits would have a population equivalent to that province’s population (13.5 million). In many studies it will be impossible to know how many people make up a population. If this is the case, it is accepted among researchers to use a fake population size of 20,000 or larger.
Your desired confidence level: Usually placed at a value of 95% in surveying, the confidence level describes how sure you can be that your results are correct. With a 95% confidence level, a researcher can be certain that the value of any sample will fall in the range of the margin of error 95% of the time.
Your allowed margin of error: Margin of error depicts the random sampling error that is possible in the study. This is important because it is impossible to know whether a sample’s results are identical with the true value of the population. The value allotted to the margin of error describes the range in value that the population may have based on the results in the study. This is always described as a plus or minus value.

For example, most people choose a margin of error 5+/- with a 95% confidence interval. If your results showed that 67% of people love rock music, you could say that you are 95% confident that 62-72% (known as the confidence interval) of your targeted population love rock music.

A Rationale for Limiting the use of First Person Pronouns for Academic Writing

Graduate students frequently wonder whether they can use first person pronouns when writing academic papers. Below is a brief guide that students can use to determine when and why writing in the first person is acceptable or not.

  1.  Review a number of articles from a highly reputable journal in your discipline. Are they written in the first person? If no, then you should model your writing in a similar style if you would eventually like to seek publication in a similar journal.
  2. Early in your academic journey, very little of what you write is truly your own work. It is most likely an interpretation of the relevant literature. Thus, there is little need to refer to yourself.
  3. When submitting a paper for a class assignment the instructor obviously knows that you wrote it and thus there is no need to refer to yourself unless stating a specific opinion. 

Are first-person pronouns acceptable in scientific writing?

 "First-person pronouns are acceptable in limited contexts. Avoid their use in rote descriptions of your methodology (“We performed the assay…”). Instead, use them to communicate that an action or a decision that you performed affects the outcome of the research."

The Scientist’s Handbook for Writing Papers and Dissertations argues that in using the third person, the writer conveys that anyone else considering the same evidence would come to the same conclusion. The first person should be reserved for stating personal opinions.

Good Style: Writing for Science and Technology2 is also against use of the first person in scientific writing, explaining that “readers of scientific papers are interested primarily in scientific facts, not in who established them.” However, this book also points out that there are points in scientific papers where it is necessary to indicate who carried out a specific action.

In Eloquent Science, Dr. Shultz concludes that “first-person pronouns in scientific writing are acceptable if used in a limited fashion and to enhance clarity.” In other words, don’t pepper your paper with I’s and We’s. But you don’t have to rigidly avoid the first person either. For example, use it when stating a nonstandard assumption (“Unlike Day and Gastel, I assumed that…”). Or use it when explaining a personal action or observation (“We decided not to include…”). Finally, follow the conventions in your field, and particularly check that the journal you intend to submit your paper to does not specifically ban the use of the first person (as a handful of journals do).

Continue reading

2016 EdTech Internship Summary

The primary objective of the EdTech internship is to provide graduate students with an opportunity to develop their understanding of the educational technology industry and educational entrepreneurship while applying their knowledge of educational research and theory to innovative technology products that seek to address the challenges of the 21st century classroom. 

This objective is achieved by providing interns with a field placement with EdTech companies that have an existing relationship with the Niagara Educational Research and Innovation Hub (ihub). 

Pairing EdTech entrepreneurs with interns that are pursuing a Masters degree in education provides these companies with the educational insight needed to ensure that their products address the needs of educational practitioners while also ensuring these products are informed by best practices and current educational research. 

For the 2016 session. MEd. student Rebecca Bunz worked with a startup creating an immersive, gamified immersive classroom platform, Answerables. Building on his experience as an ESL teacher, MEd student Keith Crawford worked with established industry leader, Strategic Transitions to support the development of resources to support second language learning.

A secondary goal of the internship is to support knowledge mobilization. As such, the interns are required to use a variety of social media resources to share their expertise and insight. Their blogs and Twitter accounts documents their experience throughout the internship while the culminating podcasts highlights their overall experience and understanding of working in the EdTech industry as well as their research pursuits and predictions for future EdTech products.

Rebecca Bunz: Twitter; Blog

Keith Crawford: Twitter; Blog

Tech-Enabled Teacher Leaders: How technology is redistributing school leadership

Similar to the way technology is transforming classroom practice, the use of technology has begun to disrupt traditional notions of educational leadership. This disruption will cause the education community to reconsider not just how school leadership is enacted, but also the role teacher-leaders play. With the plethora of tech resources that are now available, individual educators can connect, collaborate and share with educators throughout their school, district and beyond. In this article, the author examines how tech-enabled teachers from across the country are using a variety of technological resources to redistribute educational leadership and enhance the role teachers play in supporting the success of their schools and their profession.

Rutherford (2016). Tech-Enabled Teacher Leaders: How technology is redistributing school leadership. Education Canada 56(1).