Google Data Analytics Professional Certificate Process Data from Dirty to Clean Syllabus

Process Data from Dirty to Clean syllabus — full chapter and topic list

Expand any chapter to see its topics and sub-topics. This is the whole examinable outline for Process Data from Dirty to Clean in Google Data Analytics Professional Certificate, not a summary of it.

1. 1

   The Importance of Integrity

   3 topics

   • Data integrity and constraints
   • Insufficient data and limitations
   • Sample size and statistical significance
2. 2

   Sparkling-Clean Data

   3 topics

   • Identifying dirty data
   • Cleaning data in spreadsheets
   • Data validation and conditional formatting
3. 3

   Cleaning Data with SQL

   3 topics

   • SQL for large datasets
   • Cleaning string variables
   • Transforming and combining data
4. 4

   Verify and Report on Cleaning Results

   3 topics

   • Verifying the cleaning process
   • Documenting changes and the changelog
   • Reporting clean data to stakeholders
5. 5

   Adding Data to Your Resume

   2 topics

   • Highlighting analytics skills
   • Building a professional resume

Process Data from Dirty to Clean flashcards for Google Data Analytics Professional Certificate

23 of 50 cards from the Process Data from Dirty to Clean deck — real questions with worked answers.

1. What is data integrity?

   Data integrity is the accuracy, completeness, consistency, and trustworthiness of data throughout its entire lifecycle. Data with integrity is reliable for analysis and decision-making.

2. Name three ways data integrity can be compromised.

   Data integrity can be compromised through data replication (storing copies in multiple places that fall out of sync), data transfer (data lost or corrupted while moving between systems), and data manipulation (errors introduced while changing or processing the data). Human error, hacking, and system failures can also compromise integrity.

3. What is a data constraint, and what are common examples?

   A data constraint is a criterion that determines whether a piece of data is valid. Common constraints include: data type (values must be a specific type), data range (values must fall between a min and max), mandatory (value cannot be left blank), unique (no duplicates allowed), regular expression patterns (values must match a prescribed pattern), cross-field validation (conditions across multiple fields must hold), primary key (uniquely identifies each record), foreign key (values must exist in another table's primary key), set membership (values must come from a defined list), and accuracy (values must reflect reality).

4. What is data replication and why is it a risk to data integrity?

   Data replication is the process of storing data in multiple locations. It is a risk because if the copies are not updated at the same time, they become inconsistent (out of sync), leaving no single trusted version of the data.

5. What is insufficient data, and what are four common limitations of a dataset?

   Insufficient data is data that lacks what is needed to answer the business question. Common limitations: data from only one source, data that keeps updating (incomplete/still being collected), outdated data, and geographically limited data (covers only some regions when broader coverage is needed).

6. What can an analyst do when they discover they have insufficient data?

   Options include: identify trends with the available data and qualify the findings, wait for more data if time allows, talk with stakeholders to adjust the objective or the scope of the analysis, and look for a new or additional dataset.

7. In data analytics, what is the difference between a population and a sample?

   A population is all possible data values in a dataset (the entire group of interest). A sample is a part of the population that is representative of it, used when surveying the whole population is impractical or too expensive.

8. What is sampling bias and how can it be avoided?

   Sampling bias occurs when a sample is not representative of the population as a whole, so some members are less likely to be included than others. It can be reduced by using random sampling, where every member of the population has an equal chance of being chosen.

9. What is margin of error?

   Margin of error is the maximum amount that the sample results are expected to differ from those of the actual population. For example, a result of 60% with a margin of error of $\pm 5\%$ means the true population value likely lies between $55\%$ and $65\%$. Larger sample sizes give smaller margins of error.

10. What is confidence level, and what values are commonly used?

    Confidence level is the probability that the sample accurately reflects the greater population — i.e., how often you would expect to get essentially the same result if the study were repeated. Commonly used values are $90\%$, $95\%$ (the professional standard), and $99\%$.

11. What is statistical significance, and what significance threshold is commonly used?

    Statistical significance is the determination that a result is not likely due to random chance. It is commonly judged with a p-value: results with $p < 0.05$ (i.e., less than a 5% probability the result occurred by chance) are usually considered statistically significant.

12. What minimum sample size and statistical power are conventionally recommended for a reliable study?

    A conventional baseline is a minimum sample size of about 30 (based on the Central Limit Theorem, which says sample means approximate a normal distribution near this size) and a statistical power of at least $0.8$ (80%), meaning an 80% chance of detecting a true effect.

13. Which factors determine the required sample size for a study?

    Population size, confidence level, and acceptable margin of error. Higher confidence levels and smaller margins of error require larger samples; sample size calculators take these three inputs to compute the required sample size.

14. What is the difference between dirty data and clean data?

    Dirty data is data that is incomplete, incorrect, duplicated, outdated, or irrelevant to the problem being solved. Clean data is data that is complete, correct, and relevant to the problem, making analysis trustworthy.

15. List the main types of dirty data.

    Duplicate data (a record appears more than once), outdated data (superseded by newer information), incomplete data (missing important fields), incorrect/inaccurate data (complete but wrong values), and inconsistent data (the same information formatted differently across records).

16. What is a null value, and how does it differ from a zero?

    A null indicates that a value does not exist in a field — the question was never answered or the data was never recorded. A zero is an actual numeric response. Null means 'no value,' not 'value of zero.'

17. What is a field, and what does field length specify?

    A field is a single piece of information from a row or column of a spreadsheet or database (e.g., an email address column). Field length is a validation tool that specifies how many characters each field can contain.

18. What roles do data engineers and data warehousing specialists play in maintaining data quality?

    Data engineers transform raw data into a useful format for analysis by developing, maintaining, and testing databases, data processors, and related systems. Data warehousing specialists develop processes and procedures to effectively store, secure, and manage the availability of data.

19. In spreadsheets, what do the functions COUNTIF, LEN, LEFT, RIGHT, and MID do in data cleaning?

    COUNTIF counts how many cells in a range match a specified value or condition (useful for finding errors/outliers). LEN returns the number of characters in a text string (useful for checking expected lengths like ID codes). LEFT and RIGHT return a set number of characters from the left or right side of a string. MID returns a segment from the middle of a string, given a start position and length.

20. What spreadsheet tools are used to find and eliminate duplicate rows?

    The Remove Duplicates tool automatically finds and deletes duplicate entries in a spreadsheet. Duplicates can also be surfaced with conditional formatting or COUNTIF before removal. Best practice: make a backup copy of the data before removing duplicates.

21. What does the CONCATENATE function do in a spreadsheet?

    CONCATENATE joins together two or more text strings (from different cells) into a single string, e.g., combining separate first-name and last-name columns into a full-name column.

22. What does the TRIM function do and when is it used?

    TRIM removes leading, trailing, and repeated (extra) spaces from text in a cell. It is used to fix inconsistent spacing that can cause matching and sorting errors.

23. What is the difference between the SPLIT (Text to Columns) feature and the CONCATENATE function?

    SPLIT / Text to Columns divides a single text string into separate columns based on a delimiter (a character such as a comma that separates items). CONCATENATE does the opposite: it combines strings from multiple cells into one.

See more Process Data from Dirty to Clean flashcards →

Planning Process Data from Dirty to Clean for Google Data Analytics Professional Certificate

Process Data from Dirty to Clean (Google Data Analytics Professional Certificate) FAQ