Social Data Analysis

Assessment 1: Social Data Analysis
Due Date Refer to Blackboard
Length 2000 words (+/– 10%). Word count includes in-text
citations and excludes references
Weight 50%
Individual/Group Individual
Learning Outcomes • Apply a variety of data collection, curation, and
analysis software systems for the purposes of digital
media analysis.
• Apply digital methods to recognize and explain the
significance of patterns in social data.
What you need to do:
In this assignment, you will need to propose, and do some stages of, a digital media analytics
project. You will set up a topic of your choice, identify sites of data collection, collect data
from one of the sites, and discuss the interpretive implications of the dataset. There are six
steps to doing this assignment.
Step One: Topic and context (~ 300 words) No sources needed
In a couple of paragraphs, describe the context and the topic, simulating and explaining a
situation in which you will need to perform a digital data analytics project. This could be a
reflection on the career path you have chosen, a job you see yourself in, a project you have
done in a job or are going to do for another unit, or simply a hypothetical project that you
think of for this assignment. It doesn’t need to be a full-scale, big project, so please consider
the time you have for this assignment when thinking about it.
IMPORTANT NOTE
For this unit, you have received an ethical clearance to collect public social
media data and perform digital media analytics. However, since the ethical
clearance is at a unit-level, there are limitations on the topics that you can
choose for data collection and analysis. Any research about sensitive or
controversial issues, or topics that may cause harm to you or the participants
(including psychological harm) falls outside the scope of the ethical
clearance, so please make sure to carefully choose your topics.
If you are unsure about the topic you have chosen, please consult with your
tutor before collecting any data.
You can use first-person language to describe this step. For instance:
I am a data journalist working in a news agency that primarily writes for a
younger audience. We are writing a piece on the veracity of some reports
about the increasing popularity of underground doom metal music among
young people in Australia. I have been tasked with doing the research on the
conversations about this topic on social media platforms. We are mainly
interested in finding out whether this genre of music is actually increasing
in popularity, and if yes, what are some of the reasons behind it, and what
the fans of this music say about it. This requires the collection of data from
social media platforms as there are numerous communities and fan pages
discussing the topic on social media. (add more detail, e.g. about time, scope,
questions, etc.)
Step Two: Identify Platforms (~ 300 words) some sources
needed about the platforms
Reflecting on the demographics of different platforms, their norms and cultures of use,
their affordances, and the topic of the project, discuss three platforms that are best suited
for the project you are doing.
Explain what research on each platform will contribute to the project, and how each will
reveal unique patterns and insights about the questions you are interested in.
In this step, also reflect on what methods of data collection are available or necessary for
the platforms you have identified. Should you opt for a manual approach for some of the
platforms, or is there any way to use automated tools? Is scraping an option? Does the
platform have an API? Is it easy to access the API? Etc.
You can write a short paragraph (~100 words) for each platform, or use a table. Make sure
to support your arguments with references to relevant literature. Given the fast-paced
nature of changes occurring on platforms, you may find that some information is more likely
to be found in ‘grey’ literature, such as in blogs, websites, etc. It is acceptable to use these
in this section. However, please ensure you have a balance of academic and non-academic
literature. For non-academic sources, make sure to use more reliable ones.
Step Three: Collect Data
Choose one of the platforms you identified in the section above and collect the data you
need for the project, using one or more of the tools we have covered in the lectures and
tutorials. Although you may have proposed manual data collection from some platforms in
step 2, in this step you are required to showcase your skills in automated collection of social
media data,
so choose a platform that allows you to do so. Make sure to take detailed notes of how you
collected the data. You will need this information in step five.
– What tool did you use to collect the data?
– What collection parameters did you use (keywords, search terms, hashtags, dates, the
options you chose, the boxes you ticked, etc.)?
You need to submit the dataset that you collected along with your assessment.
Step Four: Investigate the Dataset
IMPORTANT NOTE
Always backup your data. Things can always go wrong. Sometimes, opening
a dataset file with some tools might change the structure of some cells. For
instance, Microsoft Excel tends to automatically convert long numbers into
scientific numbers when opening a file. This will practically make your file
unusable if saved, and you may have to collect the data again. Always have
the original file saved somewhere, make a copy, and work on the copy, so
that if something goes wrong, you still have the original file.
Open your dataset file(s) in a tool of your choice. Look at the columns and/or datapoints in
the dataset, and think about which ones are most useful for the project you have outlined in
step one. What kinds of insights can you get from each datapoint or a combination of them?
What sorts of analyses are possible? In short, what can you do with this dataset?
Your dataset might have tens of columns. You may not need all of them for the analysis, and
some of them may be more important than others for the task at hand. Reflect on these, and
decide which columns or datapoints are best to analyse for the job.
Step Five: Analytical Procedure (~1000 words) a lot of
sources needed
Once you have decided on the most relevant datapoints for the job, it is time to think
and write about how you are going to analyse them. We may have discussed some of the
approaches in the lectures and tutorials, but there are hundreds of different ways to analyse
digital media data. You may have creative and innovative ideas about how you can analyse
the dataset to obtain the insights you are after, so this step of the assignment requires some
researching and literature review, in order to find available and suitable tools that you may
need for the kinds of analyses that you propose.
In a few paragraphs, discuss how you collected the data (refer to your notes in step 3),
what columns you have chosen for the analysis, and what insights you are hoping to gain
from the analysis. How are you going to analyse them? What patterns will analysing the
dataset reveal?
How will these patterns contribute to the questions you have? What tools are you going to
use to obtain these insights?
Please note that you are not required to actually do the analysis at this stage, so don’t worry
if you don’t know how to use the tool you have identified in your search or literature review.
However, it is important that you read the relevant documents related to the tool or technique
you are proposing, to understand what can or cannot be done using it. In your writing, you
need to discuss these. Make sure to avoid using vague or broad language, and be as specific as
possible. Just as an example:
Vague: I’m going to use a programming language to find out what
proportion of accounts in the dataset are bots.
Specific: I’m going to use the Botometer API (Yang et al., 2019) in Python
to identify social bots (Ferrara et al., 2016) in the dataset. To ensure more
accuracy, the threshold set for bot identification will be 0.7.
Step Six: Reflections (~500 words) few sources needed
In a few paragraphs, discuss the different considerations associated with your analytical
process. What are the limitations of the dataset, the platform, the tools, and the techniques
you have outlined above? What about the limitations of the search parameters you have
chosen? Is it possible that you are missing something? What are the ethical considerations
associated with the project?
Submission Information
What you need to submit
A) The digital media analytics proposal that you wrote. This needs to be an Adobe PDF
or Microsoft Word document. Don’t forget to put your full name, student
number, tutorial time and day, and the name of your tutor on the first
page.
B) The dataset that you collected in step three. Submit the file in the format that the tool
you used provided (CSV, TSV, XML, Etc.). If your dataset contains more than one
file, compress them into a zip file and submit the zip file.
NOTE: On submission, you are declaring that, unless otherwise acknowledged, this
submission is wholly your work and it has not been used and already submitted. I understand
that this work may be submitted for plagiarism check and consent to this taking place.
Moderation
All staff who are assessing your work meet to discuss and compare their judgements before
marks or grades are finalised. Refer to MOPP C/5.1.7.
Academic Integrity
As a student of the QUT academic community, you are asked to uphold the principles of
academic integrity during your course of study. QUT sets expectations and responsibilities of
students specifically stating that students “adopt an ethical approach to academic work and
assessment in accordance with this policy and the Student Code of Conduct (E/2.1)”. Students
need to be aware that academic integrity refers to text and non-text sources, i.e. “copying or
adapting non-text based material created by others, such as diagrams, designs, musical score,
audio-visual materials, art work, plans, code or photographs without appropriate
acknowledgement” (MOPP C/5.3.6 Academic Integrity). It also includes self-plagiarism, this
“involves the re-use by a student of their own work without appropriate acknowledgement of
the source. Students should seek express consent from the unit coordinator prior to re-using
their own work in an assessment submission” (MOPP C/5.3.6 Academic Integrity).
Students are expected to demonstrate their own understanding and thinking using ideas
provided by ‘others’ to support and inform their work, always acknowledging the source. While
we encourage peer learning, it is not appropriate to share assignments with other students
unless your assessment piece has been stated as being a group assignment. If you do share
your assignment with another student, and they copy all or part of your assignment for their
submission, this is considered collusion and you may be reported for academic misconduct. If
you are unsure and need more information, please refer to:
http://www.mopp.qut.edu.au/C/C_05_03.jsp#C_05_03.03.mdoc.
Assignment One Marking Criteria
CRITERIA + HIGH DISTINCTION
Platform
Identification
(15%)
In-depth and thorough
understanding of platforms,
their cultures of use, and
affordances
Insightful reflections about how
investigating each identified
platform contributes to the
project
Thorough awareness of
strengths and limitations of
studying different platforms
Arguments are well supported
by authentic literature
Data Collection
(15%)
Datasets have been efficiently
collected, using appropriate
search and query parameters.
The data collection process is
very well documented and is
replicable.
Analytical
Procedure
(40%)
Highly insightful and creative
ideas proposed for the
investigation of datasets
Very suitable tools have been
identified and supported by
literature.
Proposed analysis can fully
answer the objectives set for the
project
Very clear idea of what
patterns to expect, what
datapoints to use, and how they
contribute to the analysis
Reflections
(20%)
A thorough and in-depth
understanding of the various
considerations related to digital
media data is shown.
The limitations and ethical
considerations are thoroughly
identified.
Written
Communication
(10%)
Thorough proofreading is
evident: spelling, grammar, and
punctuation are error-free
throughout.
Consistent use of discipline
specific vocabulary is adopted
throughout.
The source of external ideas is
consistently acknowledged in
thorough and correct detail in
citations and references and in
accordance with the APA style.
The document is within the
word limits.