What is the difference between data and information?
Data and information mean exactly the same thing
Data is always visual; information is always written as text
Data is stored on the internet; information is stored on a single computer
Data is raw, uninterpreted facts; information is data that has been processed and given meaning
Choose your answer
Was this exam question helpful?
What is metadata?
Data that describes, explains, or locates other data
Data that has been permanently deleted
The largest file stored on a computer
Data that has been compressed using a lossy method
Choose your answer
Was this exam question helpful?
Which of the following is an example of information rather than raw data?
A list of the individual temperatures recorded every hour
A graph showing that the temperature rose steadily through the day
A column of unsorted student test scores
A stream of GPS coordinates logged every minute
Choose your answer
Was this exam question helpful?
Raw data on its own is described as not meaningful. What must happen for raw data to become information?
It must be organised, processed, and interpreted so it is given meaning
It must be stored on a larger hard drive
It must be converted from digital form into analog form
It must be deleted once it has been collected
Choose your answer
Was this exam question helpful?
A music file stores the actual audio recording along with the artist name, album, track number, and genre. Which of these is metadata rather than the primary data?
The audio recording itself
The sound waves the speaker produces when the file is played
The artist name, album, and genre stored with the file
The headphones used to listen to the file
Choose your answer
Was this exam question helpful?
What does it mean to say that a data-processing system is scalable?
It permanently stores every file in analog form
It can handle increasing amounts of data or workload without a significant loss of performance
It reduces the size of every file it processes
It prevents any new data from ever being added
Choose your answer
Was this exam question helpful?
A shop has a long list of every individual transaction from the past year. Producing the statement "the busiest shopping hour is between 5 pm and 6 pm" from that list is an example of:
Turning data into information by identifying a pattern
Compressing the data to save storage space
Converting the data from analog to digital
Adding metadata to each transaction
Choose your answer
Was this exam question helpful?
A town observes that as the number of umbrellas sold rises, the number of car accidents also rises. What is the most accurate conclusion from this data alone?
Buying umbrellas causes people to have car accidents
The two are correlated, and a third factor such as rainy weather may explain both
Car accidents cause people to buy more umbrellas
There is no relationship between the two variables
Choose your answer
Was this exam question helpful?
What does it mean if two variables are correlated?
The two variables can never be related by any hidden factor
One variable is guaranteed to directly cause the change in the other
When one changes, the other tends to change too, but this does not prove one causes the other
The two variables are completely independent of each other
Choose your answer
Was this exam question helpful?
A user changes the recorded "date modified" field associated with a document. What happens to the document's actual written content?
The written content is permanently deleted
Nothing; metadata is independent, so editing it does not change the primary data
The content is automatically compressed
The content is converted into metadata
Choose your answer
Was this exam question helpful?
A streaming platform lets users quickly list every film released in a particular year without opening any of the film files. What makes this possible?
Converting each film from digital back to analog
Compressing every film into a single file
The release-year metadata stored with each film
Deleting the primary data of older films
Choose your answer
Was this exam question helpful?
Which two tasks can be carried out using only a file's metadata, without opening the file? Select two answers.
Reading the paragraphs written inside a document
Editing the individual pixels of a photograph
Sorting files by the date they were last modified
Filtering files by their type or size
Choose your answer
Was this exam question helpful?
Before analysis, a dataset is found to contain duplicate entries, missing values, and inconsistent spellings. The process of identifying and correcting these problems is called:
Data encryption
Data compression
Data sampling
Data cleaning
Choose your answer
Was this exam question helpful?
Two datasets are combined. One records a city as "NYC" and the other records the same city as "New York City". What kind of data-quality problem is this?
A consistency issue, because the same value is recorded in different ways
An overflow issue, because the dataset is too large
A sampling issue, because the data was measured too rarely
A compression issue, because the file was made too small
Choose your answer
Was this exam question helpful?
Why can a very large dataset ("big data") be difficult to process on a single ordinary computer?
Large datasets can only be stored in analog form
It can exceed the machine's memory or processing capacity, so it takes too long or cannot be handled
Large datasets automatically delete themselves to save space
A single computer cannot open any file larger than one byte
Choose your answer
Was this exam question helpful?
A dataset is far too large to process on one computer in a reasonable time. Which two approaches best address this? Select two answers.
Divide the work across many processors using parallel processing
Lower the resolution of the computer's screen
Convert the dataset into metadata
Use a scalable system that can expand by adding more computing resources
Choose your answer
Was this exam question helpful?
An analyst combines a shop's sales records with local weather data and discovers that sales rise on rainy days. Why can combining data from multiple sources be valuable in data analysis?
It always makes every dataset smaller and quicker to store
It converts raw data into metadata automatically
It can reveal connections that are not visible in a single dataset on its own
It guarantees that any conclusion drawn will be a proven cause
Choose your answer
Was this exam question helpful?
A town's records show that the more fire engines are sent to a fire, the greater the property damage tends to be. What is the most accurate conclusion from this data alone?
Sending fire engines causes property damage
Property damage causes more fire engines to exist
There is no relationship between the two variables
The two are correlated, and a third factor such as the size of the fire may explain both
Choose your answer
Was this exam question helpful?
Which of the following statements about metadata is true?
Metadata can only be read by opening and viewing the primary data
Metadata exists independently and can be corrected without altering the primary data
Editing a file's metadata always deletes the file's contents
Metadata and primary data are always the same size
Choose your answer
Was this exam question helpful?
A library catalog lets users search thousands of books by title, author, subject, and ISBN without handling the books themselves. This is an example of metadata being used to:
Compress each book so it takes up less shelf space
Change the text printed inside each book
Convert every book into an audio recording
Organise a large collection so items can be searched and located efficiently
Choose your answer
Was this exam question helpful?
Which two of the following are ways that computing systems use metadata?
Select two answers.
An operating system records the date, size, and type of every file it stores
A processor increases its clock speed to run programs faster
A database uses metadata to define table column names, data types, and relationships
Encryption scrambles the primary data so it cannot be read without a key
Choose your answer
Was this exam question helpful?
Before a dataset can be analysed reliably, it often needs to be cleaned. Which two of the following are data-quality problems that cleaning is meant to address?
Select two answers.
The dataset is stored on a solid-state drive
Some records are missing values in a required field
The dataset is displayed on a high-resolution monitor
The same value is recorded in inconsistent formats across records
Choose your answer
Was this exam question helpful?
An analyst notices that a sorting operation finishes in about two seconds on a small dataset but takes several hours on a very large one. What does this best illustrate?
The time required to process data scales with the size of the dataset
Large datasets are stored only in analog form
Sorting is impossible on any large dataset
The large dataset must have deleted itself to save space
Choose your answer
Was this exam question helpful?
A cloud platform analyses an enormous dataset far faster than a single processor could by using parallel processing. How does parallel processing achieve this speed-up?
It converts the dataset into metadata before analysing it
It lowers the resolution of the results to save time
It processes the data one item at a time in a strict sequence
It divides the work across multiple processors that operate simultaneously
Choose your answer
Was this exam question helpful?
A dataset shows that people who own more books tend to read faster. A researcher concludes that owning more books causes people to read faster. What is the main flaw in this reasoning?
It relies on lossy compression of the dataset
It uses too small a sample to detect any correlation
It confuses data with metadata
It assumes correlation implies causation, when a hidden factor could explain both
Choose your answer
Was this exam question helpful?
A researcher observes a strong correlation between hours spent exercising and lower stress levels. Which additional step would best justify claiming that exercise actually causes lower stress?
Carrying out a controlled experiment that rules out other explanations
Collecting the same correlated data from many more people
Displaying the correlation on a larger, more colourful chart
Assuming causation because the correlation is strong
Choose your answer
Was this exam question helpful?
A company measures customer satisfaction using an online form that only appears to visitors who have already completed a purchase. A colleague suggests collecting many more responses to make the results accurate. Why will collecting more responses not fix the problem?
Larger datasets always contain invalid data by default
More responses will make the file too large to open
The collection method itself is biased, and a biased method produces biased results at any scale
Satisfaction data cannot be stored digitally
Choose your answer
Was this exam question helpful?