Google Data Analytics flashcards
Google Data Analytics · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Data analysis lifecycle
- The six-phase framework of Ask, Prepare, Process, Analyze, Share, and Act that guides an analyst from framing a question to acting on the findings.
- Descriptive analytics
- The type of analysis that answers 'what happened' by summarizing historical data into metrics, reports, and dashboards describing past performance.
- Unstructured data
- Data with no predefined model or schema, such as free text, images, audio, and video, which needs special processing before it can be analyzed.
- Quantitative data
- Numeric data that can be counted or measured, such as revenue, click counts, or temperature, and is suitable for arithmetic and statistical analysis.
- SMART question criteria
- Specific, Measurable, Action-oriented, Relevant, and Time-bound; a checklist that turns a vague request into a testable analytical question.
- Business task
- The question or problem the analysis must answer; it defines what the stakeholder needs and what the analysis is ultimately for.
- Structured thinking
- Breaking a problem into smaller, manageable parts and mapping available data to each part before the analysis begins.
- Scope of work
- A document specifying deliverables, milestones, timeline, and resources for a project, aligning the analyst and stakeholders on expectations.
- Primary data
- Data collected firsthand by the analyst or organization for the specific project, such as surveys, interviews, or sensor readings.
- Nominal versus ordinal data
- Nominal categories have no order, such as colors or cities; ordinal categories have a meaningful order, such as small, medium, and large.
- Data dictionary
- A reference describing each field in a dataset, including its name, data type, format, allowed values, and meaning.
- Data integrity
- The accuracy, completeness, consistency, and validity of data across its lifecycle; compromised integrity undermines every downstream analysis.
- TRIM function
- Removes leading, trailing, and repeated internal spaces from text so values match correctly during sorting, filtering, and joins.
- Null value
- An empty or unrecorded value that is distinct from zero or blank text and must be excluded or handled explicitly during cleaning.
- Duplicate removal
- Finding and deleting repeated records with tools such as Remove duplicates or SELECT DISTINCT so each record is counted only once.
- Data validation
- Checking that data meets predefined rules for type, range, and format before it is used in analysis or reporting.
- Pivot table
- An interactive summary that groups rows and columns and applies aggregations such as sum or average, letting one table answer many comparison questions.
- INNER JOIN
- A join that returns only rows whose key matches in both tables, discarding unmatched rows from either side.
- LEFT JOIN
- A join that returns every row from the left table plus matching rows from the right; unmatched right-side fields appear as null.
- Correlation versus causation
- Correlation means two variables move together; causation requires evidence that one change actually produces the other, usually from a controlled experiment.
- Line chart
- A chart that displays change over time by connecting a continuous series of data points with lines, best suited to trends and seasonality.
- Pre-attentive attributes
- Visual properties such as color, size, shape, and position that the brain processes instantly and unconsciously, used to highlight key data.
- Dashboard
- A live, interactive view of key metrics that monitors current performance and updates automatically as the underlying data changes.
- Data storytelling
- Combining visuals, narrative, and context so stakeholders understand both the finding and the action being recommended.
- pandas DataFrame
- A two-dimensional labeled table in Python with rows and columns, used as the main structure for loading, cleaning, and analyzing tabular data.
- groupby aggregation
- A pandas operation that splits data into groups by a column, applies a function such as mean or sum, and returns one result per group.
- Boolean mask
- A Series of True and False values used to filter a DataFrame, keeping only the rows where the condition evaluates to True.
- pd.merge
- A pandas function that combines two DataFrames on shared key columns, with how set to inner, left, right, or outer to control retained rows.
- Capstone case study
- An end-to-end project that takes one business question through Ask, Prepare, Process, Analyze, Share, and Act, documenting decisions at each phase.
- Reproducible analysis
- Work organized so that another analyst can rerun the same steps on the same data and reach the same results.
- PII
- Personally identifiable information such as names, addresses, or ID numbers, which must be protected, minimized, and often masked before data is shared.
- Limitations disclosure
- Stating assumptions, data gaps, and uncertainty in the final report so stakeholders can judge how far the findings generalize.