Tableau Interview questions Part 2

How can we display top and bottom 5 of records in a single Tableau view?

To achieve this, we need to create 2 views View1 which contains top 5 records View2 which contains bottom 5 records View3 which is a join of View1 and View2

State few charts which we should not use with valid reasons

3D Charts: Visual representation of numbers in 3D charts will be skewed and it makes difficult to compare and analyze. Pie Charts: Pie charts are not as accurate as bar charts. In Pie charts, we have areas and angles to compare instead of length in a bar chart. Areas and angles cannot be analyzed with ease. Donut Charts: This is same as Pie chart, but here, we have a hole in the middle to make it look like donut. Due to that hole, we need to compare arc length with other arcs to analyze the values. Comparing arcs length is also a difficult task for our eyes.

Can we create cascading filters without using context filters?

Cascading filters means filter2 values are dependent on filter1 value. For Example, filter1 is country and if we select "India" for filter1, filter2 values should show all Indian states. This feature we can achieved by using option "Only Relevant Values".

What is Context Filter and show the steps on how to create the Context Filter Tableau?

Context Filters are applied to the data rows before any other filters. They are limited to views, but they can be applied on selected sheets. They define Aggregation and Disaggregation of data in Tableau Step 1: Drag the subcategory dimensions to the row shelf and the measure sales to the column shelf. Now choose the horizontal bar chart as chart type and again drag the sub category dimensions to the filter shelf. Then we will get the following chart. Step 2: Right-click on the Sub-Category field in the filter shelf and go the Top fourth tab. Choose the option field, from the next drop-down and choose the option Top 10 by Sales Sum as shown in the following screenshot. Step 3: Drag the dimension Category to the filter shelf. Give right-click on general tab to edit and under that choose Furniture from the list. As you can see the result shows three subcategories of products. Step 4: Right-click the Category: Furniture filter and select the option Add to Context. This produces the final result, which shows the subcategory of products from the category Furniture which are among the top 10 subcategories across all the products.

What is Data Modelling?

Data modelling is the analysis of data objects that are used in a business or other context and also used as identification of the relationships among these data objects. It is a first step of doing object-oriented programming.

What is data visualization?

Data visualization means the graphical representation of data or information. We can use visual objects like graphs, charts, bars and lot more. Data visualization tools provides an accessible way to see and understand the data easily.

Categorize dimensions in Tableau?

Dimensions are divided into 9 various categories Slowly ever-changing Dimension: Value of the dimension changes over an amount of time for slowly ever-changing dimensions. Example - student of worker Chop-chop ever-changing Dimension: Value in the dimension is rapidly changing for chop-chop ever-changing dimensions. Example - Age (It changes every second) Unchanged Dimension: Values are constant for unchanged dimension. Example - Traffic Signals Shrunken Dimension: Set of 1 dimension is termed as Shrunken Dimension. Example - A week is Shrunken dimension for the month Junk Dimension: Junk values or unrelated dimensions are termed as Junk Dimension. Conformed Dimension: If any dimension is provided by various business areas, then such a dimension is termed as Conformed Dimension. Example - Time (9-5) for any company or hospital or college Degenerated Dimension: Degenerated dimensions have primary keys only without any matter info. Role enjoying Dimension: If one dimension is employed in multiple roles, then they are termed as Role enjoying Dimensions. Example - Date for e-commerce site order (Date of Order, Date of Shipment, Date of delivery) Inferred Dimension: Empty dimensions are called inferred dimensions. They are usually used in ETL. Example - Customer email which he may not enter while submitting any form will be filled as null.

Explain how many types of filters are available in Tableau?

Extract Filters - Extract filters are used to apply filter on extracted data from data source. For this filter, data is extracted from data source and placed into Tableau data repository. Datasource Filters - Datasource filters are same as extract filters. They also work on extracted dataset. But, the only difference is it works with both live and extract connection. Context Filters - Context Filters are applied on the data rows before any other filters. They are limited to views, but they can be applied on selected sheets. They define Aggregation and Disaggregation of data in Tableau Dimension Filters - Dimension filters are used to apply filters on dimensions in worksheets. Dimension filters are applied through the top or bottom conditions, formula and wildcard match. Measure Filters - Measure filters are applied on the values present in the measures.

Give an overview for fact and dimension of the table?

Facts are numeric measures of data. They are stored in fact tables. Fact tables store that type of data which will be analyzed by dimension tables. Fact tables have foreign keys associating with dimension tables. Dimensions are descriptive attributes of data. Those will be stored in dimensions table. For example, customer's information like name, number, and email will be stored in dimension table.

Tell me different ways to use parameters in Tableau

Filters calculated fields actions measure-swaps changing views auto updates

State some reasons for low performance of Tableau? Explain in detail.

Filters - filters need to create an extra query and if it used in large numbers and inefficiently then they can reduce the performance. So, it is advised to use filters whenever it is mandatory Live connection - Tableau extract works much better in comparison with live connection. Data sources - a wrong query to a wrong data source can reduce performance. Also data source's performance can also affect Tableau's performance.

Differentiate parameters and filters in Tableau

Filters are the simpler and straightforward feature in Tableau. It applies to dimension or measures directly. For example, to only show Gujarat or Karnataka in a State dimension, we can apply filter on that. In Tableau, there are multiple UI options available for filters like radio buttons, drop-down lists, check boxes, sliders, and more. Filters on sheets are also available in Tableau. Parameters are like variables. They are complex and more powerful. Like a variable, a parameter can be used in calculations. So, that means, it only allows single value. Parameters have same UI options except for checkboxes because checkboxes don't have a single value. For example, we can create a parameter for interest rate and period, and then we can use these parameters to calculate interest and principal payments.

Where can we apply global filters?

Global filters can be applied to sheets, stories, and dashboards.

Differentiate between Tiled and Floating in dashboards?

In a tiled layout, items don't overlap. The layout will be adjusted according to dashboard size. In Floating layout, items can be placed on some other layers. Floating items can have fixed position and size.

How do you optimize the performance of dashboard?

Minimize the number of fields. Minimize the number of records. Reduce the marks (data points) in your view( remove unneeded dimensions). Reduce the number of filters. Use an include filter. Use a continuous data filter. Use action filters and parameters. Reduce the number of nested calculations. Remove custom SQL. Clean up your workbooks.

Which databases gives best performance when connected live?

Native Database gives best performance when connected live.

Define Page Shelf in Tableau?

Page shelf breaks the views into a series of pages. It displays an alternate view on each page. Due to this feature, you can analyze the effect of each field into the rest of the data in the view.

What is a Stacked Bar chart?

Stacked Bar Chart, composed of multiple bars stacked horizontally, one below the other. The length of the bar depends on the value in the data point. Stacked bar chart make the work easier, they will help us to know the changes in all variables presented, side by side. We can watch the changes in their total and forecast future values.

What is a Stacked Column Chart?

Stacked Column Chart, composed of multiple bars stacked vertically, one on another. The length of the bar depends on the value in the data point. A stacked column chart is the best one to know the changes in all variables. This type of charts should be checked when a number of series is higher than two.

How to download Tableau Public?

Step 1: Go to public.tableau.com and enter your e-mail address and click " Download the App ". Step 2: After downloading the file, run the file and follow the prompt to install the Tableau. Step 3: Now, you can see Tableau shortcut on your desktop. Step 4: Double click on the icon, to load tableau public. Now you see the screen as below.

Can we see SQL generated by Tableau Desktop?

Tableau Desktop log files are places in C:UsersMy DocumentsMy Tableau Repository. In case of live connection to any data source, check the log file "log.txt" and "tabprotosrv.txt" files. In case of extract connection to any data source, check the "tdeserver.txt" file which has detailed information about queries.

List out Tableau File Extensions.

Tableau Workbook (.twb) Tableau Data extract (.tde) Tableau Datasource (.tds) Tableau Packaged Datasource (.tdsx) Tableau Bookmark (.tbm) Tableau Map Source (.tms) Tableau Packaged Workbook (.twbx) - zip file containing .twb and external files. Tableau Preferences (.tps)

Are there any limitations of parameters in Tableau? If yes, give details.

Tableau dashboard allows the representation of parameters in four ways only. They don't allow any multiple values like a filter can do. They only allow a single value.

How can u set permissions or protect the data in tableau public?

Tableau does not have any security permissions or to protect the data in Tableau public. Whatever we shared in Tableau public those files, workbooks can be view by all users. There is no security option in Tableau public.

If I delete a workbook from tableau public and there are links to other blogs and other web locations. What happens to them?

Tableau public is like a repository, if you delete anything in the Tableau public the data will be lost, even if links are in other locations and blogs. We cannot access those links.

Can you get values from two different sources as a single input into parameter?

Tableau, currently not supporting the multi-valued parameters. As Tableau parameters are not dynamic, we cannot filter the list of values at runtime.

What is VIZQL in Tableau?

VIZQL is Visual Inquiry Language. It is a combination of VIZ and SQL. It is similar to SQL language. But instead of SQL commands, VIZQL language converts data queries into visual images.

Tell me something about workbook version controlled in Tableau.

Versioning of workbooks can be done in 2 ways in Tableau. At desktop level, you keep copies of your changes and files, or using 3rd party system like Microsoft TFS. At server level, where you publish the data source and version will be saved in revision history on Tableau Server or Online. Reverting to a previous version is also possible here.

89) What are the challenges faced when working with huge volumes of data?

View running will be slow. Data Extraction. Alignment issues with data. Testing Data.

How to use group in calculations field?

We can create the group by using the calculated field. Step 1: Select Analysis-> click calculated field Step 2: Enter the field name (a dialogue box will appear) Step 3: Enter the calculation -> Click Ok. Option 2: Create a set from the group, and use the set in the calculated field. Step 1: Right click on the Data Pane and then select Create Set. Step 2: Create set dialogue box will open, Enter the field name. Select the group, on general tab. Step 3: Click Ok.

Can we download views or workbooks from server? If yes, in which data formats?

We can download views or workbooks from server. But, data formats available to us depend on the permissions granted by site administrators or content owners. Image: .png format Data: .csv file. We can also download selected sheets into PDF format, but while generating PDF, web page objects won't be included.

Explain the limitation of context filters in Tableau?

Whenever we set a context filter, Tableau generates a temp table which needs to refresh each and every time the view is triggered. So, if the context filter is changed in the database, it needs to recompute the temp table, so the performance will be decreased.

Can parameters have drop down list?

Yes, Parameters have their own drop-down list, which enables the users to view the data entries which are available in the parameter during the creation.

Do we have any data limitation in Tableau Public?

Yes, Tableau Public can only allow 10 million rows to users for data visualization.

How do we do testing in Tableau?

Yes, we can do testing in Tableau by using tools and the easiest way is using the desktop application. We need to check it before we publish in the Tableau server.

We can also download selected sheets into PDF format, but while generating PDF, web page objects won't be included.

Yes, we can use unused columns in tableau filters. Let us consider one example. In data source, I have column like emp_id, emp_name, emp_sal, emp_dep, emp_designation. But, in reports, I am using emp_name in columns and emp-sal in rows. I can use emp_designation on filters.

Give a brief explanation about tableau dashboard?

is a group of various views which allows you to compare different types of data simultaneously. Datasheets and dashboards are connected, if any modification happens to the data in a sheet, then it will reflect in the dashboard. It is the most efficient approach to visualize the data and analyze it.