Tableau Data Analyst Exam

Live Connection - answer-sends queries to the database/flat file and updates the view depending on the results Extracts - answer-are saved subsets of data that improve performance and take advantage of Tableau functionality, allowing you to refresh the data on a regular basis. Published Data Source - answer-the data source can be either a live connection or an extract Flat files - answer-xlsx, csv, pdf, tdsx --.TIFF files are not accessible via tableau desktop data collection window Tableau Server - answer-customer-managed sharing platform where creators, explorers and viewers can create, interact, and share content SQL, Oracle, etc Tableau Online - answer-fully hosted (by tableau) sharing platform where creators, explorers and viewers can create, interact, and share content Tableau Prep - answer-desktop based visual ETL program. It enables developers to connect to data, create workflows, automate jobs and share flows with others. Tableau Prep - Connections Pane - answer-is on the left side and it's the space where you connect to your data sources Tableau Prep - Flow pane - answer-at the top and shows a visual representation of each step in the data prep process Tableau Prep - Profile pane - answer-displays a summary of each field in the data sample Tableau Prep - Data grid - answer-displays a preview of the rows and columns of the data source Any changes that are made in to profile pane will automatically show in the data grid Field type - answer-number, date, sting, Boolean or geographic role Alias - answer-can only be created for a dimension full vs. incremental refresh - answer-full refresh replaces all of the contents in the extract An incremental refresh only adds rows that are new since the previous refresh. logical layer - answer-the canvas for creating relationships between tables physical layer - answer-for creating unions and joins between tables (joins tables together) Aggregate - answer-rolls up the level of granularity on visible dimensions (used to reduce data size) **impacts the way your calculations function relationships - answer-create contextually appropriate joins (displayed as flexible noodles between logical tables) Joins - answer-merges multiple tables based on a clause - this creates a new fixed table Unions - answer-appends two or more tables to form a new combined table (good for data with the same number of rows and data types) Blending - answer-visualizes data from separate sources within the same view via blended fields (the result is similar to a left join) Specific and Wildcard - answer-types of unions that are available in the physical layer Dimensions - answer-qualitative (used to categorize, segment, and filter) eg names and dates (normally discrete) Measures - answer-quantitative (can be counted or aggregated) eg $, ratios, or quantities (normally continuous) Discrete - answer-(blue) - finite set of values -> year, category, country ADD A NEW HEADER (normally dimensions) Continuous - answer-infinite range of values -> age, temperature, profit -> ADD A NEW AXIS (normally measures) Bar Charts Line charts Histogram - answer-Require 1 measure and 0 dimensions Scatter plots Multi-axis charts - answer-Require 2 measures and 0 dimensions Crosstabs Highlight tables Heatmap - answer-Require 1 measure and 1 or more dimensions Tree maps - answer-Require 1 or 2 measures and 1 or more dimensions Bullet Graphs - answer-Require 2 measures and 1 or more dimensions Geographic Fields and Spatial Data Files - answer-The two primary forms of geospatial data View Toolbar - answer-map workspace feature gives users the ability to pan and zoom Census data - answer-is already built in to tableaus data layer Geographic fields - answer-assign a geographic role. Most will auto assign. All are string type data except for latitude and longitude Special data fields - answer-these use geometry fields. This is usually a special .shp file. When dropping a Geometry field, everything will display as one mark. zip codes - answer-should be coded as strings to avoid map errors Analytics Pane - answer-Find the options to add a: Trend Line, Forecast, Custom Reference Line, Reference Band, Distribution Band, Box Plot Regression Models - answer-In tableau they are represented in the form of trend lines It is a statistical modeling technique used to estimate trends/relationships between variables Logarithmic - answer-When a trend increases then levels off (only positive values) Linear - answer-shows the relationship between variables (either increasing or decreasing) Exponential - answer-the rate of change rises at increasingly higher rates (positive only) Polynomial - answer-when variables have a fluctuating relationship (revenue over time) Power - answer-When the dependent variable (y) changes at a predetermined rate R-squared - answer-explains how the variance of one variable explains the variance of another. R2 ranges from 0-1, and closer to 1 is better. P value - answer-· is the probability of obtaining results as extreme as observed results. P value ranges from 0-1, and a value of <.05 is good. It means the trendline is an accurate predictor of future trends. Forecasting - answer-algorithms evaluate patterns in historical data to a project potential future data point. -> drag and drop forecast tools from the analytics pane -> choose forecast length order of operations tableau uses to calculate - answer-1. Extract Filters 2. Data Source Filters 3. Context Filters 4. Dimension Filters 5. Measure Filters 6. Table Calc Filter Calculated Fields can - answer-Segment data, convert field types, aggregate data, filter results, calculate new metrics Aggregation - answer-Combining and summarizing data into a single representative value. Both dimensions and measures can be aggregated Measure calculations - answer-sum, average, median, count, Min/max, Stand. Dev., variance Dimension calculations - answer-min/max (a/z), count, distinct count LOD Expressions - answer-type of calculated field leverages computations that support aggregation at dimensionalities other than the view-level supports aggregation at dimensions other than view level unless this is used, calculations will always compute at the level of detail shown in the view LOD element - exclude - answer-compute at a coarse grain or higher than the view (impacted by filters/dimensions in the view) LOD element - include - answer-compute at a finer grain and are lower than the view (impacted by filters/dimensions in the view) LOD element - fixed - answer-compute at a course or fine grain. They are not impacted by filters/dimensions in the view Types of aggregation - answer-1. Row-level aggregation - performed at the lowest level of the dataset example: [Profit] / [Sales] 2. Dynamic Aggregation - aggregation at the level of detail in the view example: SUM[Profit] / SUM[Sales] 3. User defined aggregation - aggregation at the level of detail explicitly defined in LOD expression. [FIXED [city], [state]: ([sales])) aggregation mix error - answer-when aggregate and non-aggregate fields are used in a calculation String - answer-allows for the manipulation of text-based data Contains, Len, Replace, Split, Trim, upper/lower, regex_replace, regex_match COUNTD - answer-converts a dimension into a measure