GUIDEBOOK

RMIT BUSINESS ANALYST CHAMPION 2020

Contents.

03
Part A: Data Analytic Process

25
Part B: Presentation

27
Part C: Data Analytic Tools

02
RMIT BUSINESS ANALYST CHAMPION 2020

PA R T

A
Data Analytic Process

03
 RMIT BUSINESS ANALYST CHAMPION 2020

DATA ANALYTIC PROCESS

04
 RMIT BUSINESS ANALYST CHAMPION 2020

STEP 1: DEFINE OBJECTIVES AND INDENTIFY TOPIC

1. CREATIVE PROBLEM SOLVING PROCESS

Creative Problem Solving may be defined as a problem solving technique that

addresses a challenge or problem in a creative manner. The solution is creative be-
cause it is not obvious. To meet the criteria for solving a problem in a creative manner,
the solution should resolve the declared problem in an original manner with the solu-
tion being reached independently. This idea generation strategy usually incorporates a
team approach.
There are six stages in the Osborn-Parnes process of creative-problem solving:

05
 RMIT BUSINESS ANALYST CHAMPION 2020

2. OBJECTIVE-SETTING METHODOLOGY

a. BSQ
“BSQ” stands for big, small, and quick:
• Think Big: Define your ultimate goal.
• Act Small: Identify the milestones that will help you achieve your goal.
• Move Quick: Come up with a timeline for achieving each milestone.

b. SMART

SMART is an acronym that explains the characteristics that contribute to effective goals:

06
 RMIT BUSINESS ANALYST CHAMPION 2020

• Specific: Your goal should be precise. Rather than setting the general goal to read
more, decide how often and how much you would like to read.

• Measurable: Your goal’s failure or success can be calculated in some way. For
example, you can measure your progress by the number of pages you have read in a
day or week, or by the amount of time you have spent reading in a day or week.

• Attainable: Your goal should stretch your abilities but still remain possible. When
you set an achievable goal, you may be able to identify previously overlooked oppor-
tunities or resources that can bring you closer to it.

• Relevant: When setting goals for yourself, consider whether or not they are rele-
vant. Each of your goals should align with your values and larger, long-term goals.

• Time-based: Your goal should have a deadline. Give yourself neither too much nor
too little time to accomplish your goal. A goal without a deadline will almost certainly
be.

07
 RMIT BUSINESS ANALYST CHAMPION 2020

c. OGTM
The OGTM (or objectives, goals, tactics, and metrics) method for goal-setting starts
broad and becomes specific:
• Objectives: First, create an objective. The OGTM method defines an objective as a
generalized, long-term hope that may be accomplished in a few months or even a few
years.
• Goals: Streamline your objective into a goal (or a more tangible target).
• Tactics: much like the “act small” step in BSQ, refer to the techniques that enable
you to reach your objective.
• Metrics: much like “measurable” from the SMART acronym, refers to the method-
ology for assessing a goal’s success. This step relies on numbers (such as percent-
ages) to tell a story about the extent to which the objective was achieved.

***Although BSQ, SMART, and OGTM contain differ-

ent letters, they embody similar ideals. No matter which
goal-setting framework you choose, make sure your goals
are specific, measurable, and attainable

08
RMIT BUSINESS ANALYST CHAMPION 2020

STEP 2: COLLECT DATA

09
 RMIT BUSINESS ANALYST CHAMPION 2020

STEP 3: DATA ANALYSIS AND DATA VISUALIZATION

1. DATA ANALYTICS
Business analytics is a multi-stage process. Each step involves the analysis of data to
reach a particular type of conclusion, the ultimate goal of which is to build the best possi-
ble strategy for optimized organizational action.
Types of Data
• Quantitative: containing variables that describe quantities of the objects of interest.
The values are numbers. The weight of an infant is an example of a quantitative vari-
able.
• Qualitative: containing variables that describe qualities of the objects of interest.
These values are called categories, also referred as levels or modalities. The gender
of an infant is an example of a qualitative variable. The possible values are the cate-
gories male and female.
• Mixed: containing both types of variables.’

a. Descriptive Analytics
(DEA)
This branch of Business Analytics analyses and finds an answer to the question “What
has happened in the past”. It’s all about using a range of historic data to draw compari-
sons, summarizing raw data, statistics so that it is easily understandable.
DEA best practices
• Is rule-based and more directly
• Process data into results readable at a glance

10
 RMIT BUSINESS ANALYST CHAMPION 2020

b. Exploratory Analytics
(EDA)
Exploratory Data Analysis refers to the critical process of performing initial investigations
on data so as to discover patterns,to spot anomalies,to test hypothesis and to check as-
sumptions with the help of summary statistics and graphical representations
EDA best practices
• Have a goal or a problem to solve
• Do a controlled exploration
• Frequently review your finding and direction
• Do not get diverted.

c. Explanatory Analytics
(EPA)
EPA deals with the question “Why did it happen?” The goal of explanatory analytics is
to identify reasons and root causes for business results. Explanatory analytics seeks to
tell stories with data. It is about following a methodical process of starting with a ques-
tion, exploring through data, and finding an end answer.
EBA best practices are the same with EDA, but added following practices:
• Know your audience(background, skill levels, expertises, etc,...), prepare your pre-
sentation accordingly
• Make sure your presentation flows logically from the question to the answer
• Do not let the audience distract the meeting by sidetracking into non-relevant
• Confirms that the cause identified by the analytics is actually true and not something
like missing data.

d. Predictive Analytics
Predictive analytics can also be called predictive modeling. Basically, it is away of match-
ing data with predictive models and defining a likely outcome.
Best Practices:
• When you try to use a model for actual business purposes, focus on the business
gain or return on investment, not just model accuracy.
• Focus on data and getting more data
• Test model building with multiple algorithms and see which one fits best for your use
case based on accuracy and response times.
• Ensure that all relevant variables that impact the outcome are considered during pre-
diction.
• Test, retest, and periodically test model accuracy with new sets of data

11
 RMIT BUSINESS ANALYST CHAMPION 2020

***Predictive analytics teams should have the right com-

position of talent including data engineering, statistics,
machine learning, and business.

e. Prescriptive Analytics

Prescriptive analytics builds on predictive analytics by helping determine recommended

(prescribed) actions based on desired potential (predicted) outcomes, helping organiza-
tions achieve their business objectives. In the wider view of applying business analytics
for organizational success, prescriptive analytics “delivers business value through
recommendations” built on the data of results.

Prescriptive Analytics best practices

• Provides tangible, measurable benefits

• Supports what-if scenarios

• Accounts for all inputs, variables and outputs

• Learns from transactions made in past

• Can be applied to a vast number of decisions

• Identifies what’s likely to happen

• Presents the user with specific options and determines which option are best in terms
of certain criteria

12
 RMIT BUSINESS ANALYST CHAMPION 2020

2. DATA VISUALIZATION
After preparing the data sets, we would visualize them for better analysis, using Excel’s
different types of charts and following the rules of data visualization for better readers’
experience.

a. Types

13
 RMIT BUSINESS ANALYST CHAMPION 2020

***Six data dimensions (X-axis placement, Y-axis

placement, Size, Shape, Colour, Animation (in-
teractive only, often used to display time)

14
 RMIT BUSINESS ANALYST CHAMPION 2020

b. Six meta-rules
for data visulization

01 02
Simplicity over Complexity Direct representation
The simplest chart is usually the one Always directly represent the relation-
that communicates most clearly. Use ship you are trying to communicate.
the “not wrong” chart – not the Don’t leave it to the viewer to derive
“cool” chart. the relationship from other informa-
tion.

03 04
Single dimensionality Use colour properly
Generally, do not ask viewers to com- Never use colour on top of colour –
pare in two dimensions. Comparing colour is not absolute.
differences in length is easier than
As with shape, colour should be used
comparing differences in area. to provide the viewer meaning. Hence,
it should be used consistently, and
within several rules for human percep-
tion.

05 06
Use viewers’ experience Represent the data story
to your advantage with integrity
Do not violate the primal perceptions Chart with graphical and ethical in-
of your viewers. Remember, up means tegrity. Do not lie, either by mistake or
more and the maximum percent level intentionally.
is 100%, etc.
It is important that, at all times, the
When a visualizer violates these per- effect in the data must be accurately
ceptions, it can cause cognitive confu- reflected in the visualization.
sion in the viewer and create a nega-
tive emotional reaction in the viewer.

15
 RMIT BUSINESS ANALYST CHAMPION 2020

2. CREATE AND ORGANIZE DASHBOARDS

Dashboards contain multiple worksheets which are linked. Hence, the action in any of the
worksheet can change the result in the dashboard accordingly.

AT TA C T I V E DA S H B O A R D T I P S
a. Advoid clutter

Remove as much as you possibly can while ensuring the end user gets the right insight
from your dashboard.

b. Use a grid layout

Use a columnar- or row-based flow to create a narrative that leads users from overview to
detail.

16
 RMIT BUSINESS ANALYST CHAMPION 2020

c. Simplify your use of color

AT TA C T I V E DA S H B O A R D T I P S

You should be able to justify every single color on your dashboard: why did you choose
any specific color, and what does it communicate to your user? If you can’t answer that
question, remove the color.

d. Storytelling
After dealing with numbers and charts for the dashboard, you should have a good story
to convey your ideas to others. Simply, you have to create story points and plan their
orders together with the dashboards you have created before. Make sure to size your
dashboard correctly for the story.

17
 RMIT BUSINESS ANALYST CHAMPION 2020

STEP 4: Decision Making

PROBLEM SOLVING METHODOLOGY

a. Traditional thinking
pattern
When solving problems people search for the answer to the following four questions:
• What happened?
• Why did it happen?
• How should we act?
• What will be the (future) result?

b. Kepner-Tregoe method
(KT-method)
The Kepner Tregoe method is a problem analysis model in which the “problem” is
disconnected from the “decision”.
To breakthrough traditional thinking pattern, Charles Kepner and Benjamin Tregoe
came up with four rational processes in which four fundamental questions are reflect-
ed:
• Situation analysis
• Problem analysis
• Decision analysis
• Potential problem analysis

18
 RMIT BUSINESS ANALYST CHAMPION 2020

c. Rational Problem-solving
A person with this preference often prefers using a comprehensive and logical ap-
proach similar to the guidelines in the above section. For example, the rational ap-
proach, described below, is often used when addressing large, complex matters in
strategic planning.
• Define the problem.
• Examine all potential causes for the problem.
• Identify all alternatives to resolve the problem.
• Carefully select an alternative.
• Develop an orderly implementation plan to implement that best alternative.
• Carefully monitor implementation of the plan.
• Verify if the problem has been resolved or not.
A major advantage of this approach is that it gives a strong sense of order in an other-
wise chaotic situation and provides a common frame of reference from which people
can communicate in the situation. A major disadvantage of this approach is that it can
take a long time to finish. Some people might argue, too, that the world is much too
chaotic for the rational approach to be useful.

d. Hypothesis-driven solving
The most fundamental skill of any problem-solver is hypothesis-based thinking. Struc-
turing analyses in this way ensures efficiency and focus because only relevant data
is being analyzed. A crucial part of the hypothesis-based methodology is to ensure
that the formed hypothesis is falsifiable. When a falsifiable hypothesis is formed, one
should construct a framework that enables testing of the hypothesis and collect the
data required to do so.

19
 RMIT BUSINESS ANALYST CHAMPION 2020

d. The MECE issue tree

Issue trees are used to break down complex problems into more manageable sub-is-
sues. While the previous section explained the process of hypothesis testing, the
MECE framework helps identify the relevant hypotheses. MECE is an acronym for
“Mutually Exclusive, Collectively Exhaustive” and is a way of organizing informa-
tion - often in the context of issue trees. If the issue tree is done in a “MECE” way,
all sub-issues collectively exhaust the solution space (all possible solutions) and the
sub-issues are mutually exclusive (no sub-issues are overlapping). A MECE issue tree
ensures that no possible solution is missing and that work streams can be distributed
among the case team members without any duplication of the work.

20
 RMIT BUSINESS ANALYST CHAMPION 2020

e. Synectics
This problem solving methodology inspires thought processes that the subject might not
be aware of. In Gordon’s opinion, Synectics research has to do with three key assump-
tions:
• It is possible to describe and teach the creative process;
• Invention processes in science and arts are analogous and propelled by the same
“psychic” processes;
• Creativity at the level of individual and group is analogous.
In short, if people comprehend the working of creativity, they can improve their ability to
be creative.

f. TRIZ methodology
TRIZ (or TIPS – Theory of Inventive Problem Solving) is a Russian method of problem
solving. This strategy is meant to cultivate the creation of patentable inventions. Howev-
er, the technique is also helpful for developing non-product solutions.

21
 RMIT BUSINESS ANALYST CHAMPION 2020

g. Reversal of problem

This approach is about coming up with ideas to solve problems by way of a different/op-
posite perspective (turning it around: upside-down, inside-out or back to front).

EXAMPLE

22
 RMIT BUSINESS ANALYST CHAMPION 2020

h. Look beyond something’s

common function
Split an object into all its individual parts. If you have a description suggesting a function
(just like the function of a prong is transporting electricity), describe it in a more generic
manner by way of shape, size and the make-up of the material (such as rectangular, flat,
small piece of metal). If you call an item an electric plug’s prong, the description may
conceal the fact that the item could also turn into a screwdriver if required.

i. Lateral thinking

Lateral thinking is a manner of thinking that looks for a solution to an obstinate issue
through unorthodox elements or methods that would usually be disregarded by logi-
cal thinking. To be more precise, “lateral thinking” may be defined as a way to solve
problems by a creative or indirect approach, utilizing reasoning that may not be obvious
straight away or incorporating ideas that cannot be gathered by utilizing only convention-
al step-by-step logic.

Challenge, alternatives, and provocation and movement are three examples of lateral
thinking techniques.

j. SCAMPER

23
RMIT BUSINESS ANALYST CHAMPION 2020

PA R T

BP resentation

25
 RMIT BUSINESS ANALYST CHAMPION 2020

Every slide should serve one of the following three purposes:

• Communicate a set of key implications or facts.
• Guide the audience through different analyses or create a logical transi-
tion.
• Drive towards recommendation. Below we give an example of how to
structure a slide.

1. Slide writing commandments:

• Synthesis: Synthesize many analyses into one story
• Accuracy: Use precise, clear communication with no ambiguity
• Transparency: Transparent logic and connection between data and conclusion
• Impact: Highlight key implications and explain the insights
• Context: Tell the story that addresses your audience’s concerns, not your story
2. Design principles:
• Show it, don’t tell it - a figure is worth a thousand bullets
• Less is more - design your slides for quick reading
• Keep focus on what is important - avoid excessive use of colors
• Focus on one key message per slide
3. Text of graphs?
• Qualitative information → Text slide with supporting visuals
• Quantitative information or relationship → Graph
• When details are demanded →Table or matrices
• Clear graphics to support conclusion, explicit text to explain insight

26
RMIT BUSINESS ANALYST CHAMPION 2020

PA R T

C
Data Analytic Tools

27
 RMIT BUSINESS ANALYST CHAMPION 2020

SUMMARY STATISTICS

1. MEAN (X): the average of a data set.

Formula:

2. THE MEDIAN: is the middle value of an ordered

sample of a data set

To calculate the median:

• Arrange all the recorded values in order of size
• Find the middle value:

3. MODE

• The mode is the value which occurs most often in the group. This can be a group of
either numbers or categories.
• If there are two modes in the group of numbers, the group is described as bimodal.

28
 RMIT BUSINESS ANALYST CHAMPION 2020

4. STANDARD DEVIATION

• Measures the average distance between a single observation and the mean
• The larger the standard deviation, the more the values differ from the mean, and
therefore the more widely spread out.

Formula:

5. STANDARD ERROR
The estimated standard deviation of the population mean. It shows how precise the
model’s predictions are by demonstrating how far the data points are from the regres-
sion curve on average. The lower the better, because it signifies that the distances
between the data points and the expected values are smaller
Formula:

29
 RMIT BUSINESS ANALYST CHAMPION 2020

6. 95% CONFIDENCE INTERVAL FOR THE MEAN (95%

LCL & 95% UCL)

This is the upper and lower values of a 95% confidence interval estimate for the mean
based on a t distribution with n-1 degrees of freedom.
Formula:

7. RANGE

The difference between the largest and smallest data values.

Formula:
Range = Maximum - Minimum

8. PERCENTILES
The 100pth percentile is the value below which 100p% of data values may be found
(and above which 100p% of data values may be found).
The 100pth percentile formula:

9. INTERQUARTILE RANGE (IQR)

The difference between the 75th and 25th percentiles (the 3rd and 1st quartiles). This
represents the range of the middle 50% of the data and serves as a robust measure of
the variation in the data.
Formula:

30
 RMIT BUSINESS ANALYST CHAMPION 2020

10. MEAN ABSOLUTE DEVIATION (MAD)

A measure of dispersion that is not affected by outliers as much as the standard devia-
tion and variance.
Formula:

11. MEAN ABSOLUTE DEVIATION FROM THE MEDIAN

(MADM)
• A measure of dispersion that is even more robust to outliers than the MAD.
Formula:

12. COEFFICIENT OF VARIATION (CV)

• A measure of the dispersion of data points in a data series around the mean.
• It can be used to compare two or more data sets in different units or in different
means.
Formula:

31
 RMIT BUSINESS ANALYST CHAMPION 2020

13. SKEWNESS

Measures the direction and degree of asymmetry in the data distribution.

Formula:

14. KURTOSIS

Measure the heaviness of the tails in the data distribution.

Formula:

32
 RMIT BUSINESS ANALYST CHAMPION 2020

MULTIPLE REGRESSION MODEL

1. MODEL AND INTERPRETATION

• Application: Examining the relationship between 1 dependent (Y) and 2 or more

independent variables (Xi) to estimate Y value from the values of X
• Multiple regression model:

• Multiple regression equation:

Coefficients of the multiple regression model are estimated using sample data:

33
 RMIT BUSINESS ANALYST CHAMPION 2020

2. GOODNESS-OF-FIT

• Coefficient of Determination (R2)

Shows the percentage of the variation in the dependent variable that can be explained
by the variation in the included independent variables, hence the higher the better.
However, R2 does not care about the insignificant points. Even adding insignificant
variables can make it increase, which may lead to biased prediction.

• Adjusted Coefficient of Determination (Adjusted R2)

An adjustment for the Coefficient of Determination that takes into account the number
of variables in a data set. Still following the higher the better, Adjusted R2 will only
increase if a relevant and significant variable is added, showing that the regression is
actually improved.

3. RECOGNIZING SIGNIFICANT VARIABLES

Use Hypothesis testing for each and every individual variable slope to determine which inde-
pendent variables have a statistically significant relationship with the dependent vari-
able. If a variable is statistically significant, discuss the magnitude of the coefficient to
get an idea of its economic or practical importance. However, keep in mind the fact that
a coefficient is statistically significant does not necessarily mean it is economically or
practically significant.

34
RMIT BUSINESS ANALYST CHAMPION 2020

35
RMIT BUSINESS ANALYST CHAMPION 2020

36
 RMIT BUSINESS ANALYST CHAMPION 2020

EXCEL

1. DATA PREPARATION
After importing and Before visualizing data, we need to initially clean and prepare
them. Using the following functions and shortcuts in Excel would make the job easier
and more efficiently.

37
RMIT BUSINESS ANALYST CHAMPION 2020

38
RMIT BUSINESS ANALYST CHAMPION 2020

39
 RMIT BUSINESS ANALYST CHAMPION 2020

2. DATA ANALYSIS TOOLS

Besides, Excel also have tools to help us with analysis such as regression or descrip-
tive statistics:

40
 RMIT BUSINESS ANALYST CHAMPION 2020

Pivot table
A pivot table is a data summarization tool that is used in the context of data process-
ing. Pivot tables are used to summarize, sort, reorganize, group, count, total or aver-
age data stored in a database. Below are some of the tools you may need when using
pivot tables:

a. Insert/Remove subtotals
and grand totals
Subtotal is the sum of a set of numbers, which is then added to another set(s) of num-
bers to make the grand total. The Subtotal feature is not limited to only totaling subsets
of values within a data set. It allows you to group and summarize your data using SUM,
COUNT, AVERAGE, MIN, MAX and other functions.

b. Summarize values by/

Show values at:

41
 RMIT BUSINESS ANALYST CHAMPION 2020

c. Grouping

You can group numbers in Pivot Table to create frequency distribution tables. This
helps in analyzing numerical values by grouping it into ranges.

d. Sorting

To help you locate data that you want to analyze in a PivotTable more easily, you can
sort text entries (from A to Z or Z to A), numbers (from smallest to largest or largest to
smallest), and dates and times (from oldest to newest or newest to oldest).

42
 RMIT BUSINESS ANALYST CHAMPION 2020

e. Filtering

There are several methods for filtering data: insert one or more slicers, apply filters
to any field in the PivotTable’s Row field with AutoFilter, add filters to the PivotTable’s
Filter field.

f. Pivot charts

A pivot chart is the visual representation of a pivot table in Excel. Pivot charts and pivot tables
are connected with each other.

43
Design and Edit
Le Minh Hanh
RMIT Bachelor of Professional Communication

Support and Edit

Vo Thi Hong Nhung
RMIT Bachelor of Profressional Communication