Author: Shibani

Competency 2.3

Competency 2.3: Evaluate the impact of policy and strategic planning on systems-level deployment of learning analytics.

#Assignment 68

Learning analytics could create a bigger impact on learning if implemented top-down than bottom up due to the availability of big data.However, the deployment of learning analytics faces many challenges at the institutional level:


1. Acceptance:
To work big on big data, big support is needed from the top management. The top management should foresee the future and possibilities of learning analytics as to what it can achieve. Only with promised outcomes, they can be expected to support it at a big level. It is not a small change to bring about in a day.

2. Management:
A new department may be needed to manage what should be done in learning analytics. This will require funding, responsible experts, manpower and technical training. Do the institutions have what it takes to commit to this new venture?

3. Ethics:
Personal Data Protection is a growing concern these days. When data is analyzed, it has to pass through humans and systems. How safe can our data be? Could there be a possible breach in security and what could be its implication?

When we have answers for all these questions, we could probably move forward to the next era of data analytics!


Competency 2.2

Competency 2.2: Download, install, and conduct basic analytics using Tableau software.

For the purpose of this competency, I’ve used a sample excel sheet with simple sales data. I’ve installed Tableau for this exercise and used a two sheet excel. My sample data below:


I’ve then added rows and columns to create visualizations. Common tools from the Show me list are used to visualize my analysis of Sales by Rep and Sales by Region.

I’ve then created a dashboard with my charts.

#Assignment 47

Competency 2.1

Competency 2.1: Describe the learning analytics data cycle.

Learning Analytics Data Cycle
The process of Learning Analytics is a cycle and not linear due to the fact that we need to revisit the steps according to the data and results.

I see the data cycle as follows:


Cleaning the data could be time consuming and challenging depending on the nature of data. If the data is not cleaned thoroughly, it will not produce accurate results. For example, a data set where duplicates are not removed when processed will not yield expected results.


Manipulating the data will help us make data easier to work with. We can change the set of data to deal with, reshape it to another form, reorder the data etc. according to our requirements.

Analyzing the data is to make sense of the data using different methods. It requires adding in possible transformations to process the data. Based on the analysis, prediction models can also be built for future data.

Viewing the data involves looking into the processes data/ results using various visualization techniques. A lot of visualization options are available to try and understand data.




Competency 1.2

Competency 1.2: Define learning analytics and detail types of insight they can provide to educators and learners.

My definition of Learning Analytics (based on previous definitions and my understanding):

Learning Analytics is something which will help us to process simple data into useful information using different methods. A lot of minute, but useful information may be lost in large amounts of data, which can be extracted using learning analytics. The different methods can help us to convert raw data to a readable form, analyze using required processing and view results using possible visualizations. 

Educators and learners can get a variety of insights using learning analytics such as:
Statistical Analysis
Discourse Analysis
Text processing
Sentiment Analysis
Network Analysis
Prediction models
Machine Learning
Visualization techniques

The list is not exhaustive. We could also add specific examples to the list.

Competency 1.1

Competency 1.1: Identify proprietary and open source tools commonly used in learning analytics.

 I have specifically focused on text mining while searching for tools because I’ve been working on it, but some of these also include other options. From my usage, the most powerful ones I’ve seen are Rapidminer and R, both available for free. 

Tool
Publisher
Reference
Website
Text Analysis Methods
WORDSTAT
Provalis Research
Word Categorization,
Frequency Analysis,
Keyword retrieval,
Automated text classification
TagHelper
Carolyn Rose
Automatic coding based upon the written coding rules
MEPA
Gijsbert Erkens
frequency, time-interval analysis,
 word-frequency,
 word-context analysis, sorting and searching
SPSS Text Analytics
IBM
Automatic categorization and grouping of terms
Wmatrix
Paul Rayson
Word frequencies, word search, word clouds, pos tag frequencies, MWEs, n-grams, automatic tagging
KNIME
Kilian Thiel
classification and clustering of documents, named entity recognition, tag clouds
Weka
The University of Waikato
Text classification, clustering, association
TADA-Ed
Agathe Merceron,
Kalina Yacef
http://imej.wfu.edu/articles/2005/
1/03/index.asp#2
Classification, clustering, association
RapidMiner
RapidMiner Inc
Operator pane –> Text Processing folder. There are several more folders such as “Tokenization”, “Extraction”, “Filtering”, “Stemming”, “Transformation”, and “Utility”
R
The R Foundation for Statistical Computing
frequent terms, clustering, classification, association analysis

Orientation notes and Competency

NOTES

DALMOOC Course Basics:

DALMOOC emphasizes on Connections (learners) and Creation (Artifacts) and not much on content. It aims to build upon Networked knowledge and Combinatorial Creativity. It encourages users to own their learning space, so that the knowledge does not perish in discussion forums, comments etc.

Distributed environment:
The contents and conversations can be distributed in various environments:
Blogs
Twitter and social media
ProSolo
EdX

Content flow:
In normal teaching environment, the content flow is from the instructor to the student; whereas in DALMOOC, the core content is co-created by faculty, learners and external experts.

Key Activity:
Create/ Share
Revise/ Remix

Wayfinding and Sensemaking:
Visual syllabus
Follow daily email
Hangouts or recordings
edX forums

Orientation Hangouts:
George, Carolyn, Matt, Dragan, Ryan

Feedback:
Experimenting with social learning, new tools, course content etc. So, feedback from learners expected.

Quickhelper:
To get guidance from helpers, post a discussion in Quickhelper and someone will respond to it.

ProSolo:
To document activities and develop competencies

Collaborative activity:
Synchronous chat when partner is available.
Lobby connects to partner for chat
Agent leads each step when ready

Assessment:
No peer assessment; self assessment only by pasting evidence for competency.

Course Agreement:
Provide consent for participation in the DALMOOC Course Agreement,


COMPETENCY 0.1 Assessment

Competency 0.1: Describe and navigate the distributed structure of DALMOOC, different ways to engage with peers and various technologies to manage and share personal learning.





Introduction

It’s been four weeks since the Data, Analytics and Learning Massive Online Course started on 20th October 2014. This is my first online course and I’m sad that I missed participating in the first few weeks. Nevertheless, I still have hope since I was able to sign up in Mid November and I’m confident that I’ll be able to finish the course successfully. 

In the next few days and weeks, I will be posting my reflections and competencies for assessment. I just started a learning marathon today, aiming to learn what I should have learnt in weeks in a few days. I know its late, but better late than never!

#DALMOOC