Study 8. From Data to Insight with 12 free online flashcards. Review key terms, definitions, and concepts with this interactive flashcard deck.
What workflow turns data into reliable insight?
A reliable workflow defines a question, collects appropriate data, represents and cleans it, explores patterns, visualizes results, interprets evidence, and communicates conclusions responsibly.
In a dataset, what do rows and columns represent?
Rows represent observations, such as sales or participants; columns represent variables, such as date, price, location, or age.
What are a bit and a byte?
A bit is a binary digit with a value of 0 or 1. A byte contains 8 bits and can represent 256 different patterns.
What decimal number is 10110₂?
22₁₀. Calculate 1×16 + 0×8 + 1×4 + 1×2 + 0×1.
Why can the same bits mean different things?
An encoding rule assigns meaning to bit patterns. The same bits may represent text, image color values, or other information depending on the format.
How do lossless and lossy compression differ?
Lossless compression preserves the exact original data; lossy compression produces smaller files by discarding information that may be less important for the intended use.
Why should missing data not be converted silently to zero?
A blank may mean not collected, declined, or not applicable, whereas zero means a quantity was measured and found to be zero.
How should an outlier be handled?
Do not automatically delete it. Investigate whether it is an error, sensor problem, rare legitimate event, or process change; document the decision and its effect.
What do filtering, sorting, and grouping do?
Filtering selects records meeting conditions, sorting orders records, and grouping combines records into categories for comparison or summary.
Which charts fit common analytical questions?
Use a line chart to show change over time, a bar chart to compare categories, a histogram or box plot to show distributions, and a scatter plot to examine two numeric variables.
Why does correlation not establish causation?
Correlation shows that variables change together; it does not by itself prove that one causes the other. A confounding variable, coincidence, or reverse influence may explain the association.
Why compare rates instead of totals?
Use an appropriate denominator, such as incidents per 1,000 people or sales per store. Totals can be larger simply because the population or number of stores is larger.