Study guide: Design Data that Answers a Question

Ideas to explain in your own words

• A field is a piece of information recorded for each item. Useful fields depend on the question. To compare toy materials, material and toy type matter; a home address does not. Collecting fewer irrelevant details can make a record easier to use and less intrusive.
• A data format tells how to record values consistently. Different spellings can split one category into several apparent groups. A dictionary or key can define allowed values. Our lesson uses a tiny codebook so a learner can normalize categories without guessing missing information.
• A record should represent the stated unit, such as one item or one observation. Repeated item IDs may indicate accidental duplicates, but repeated observations can also be legitimate when dates differ. The supplied policy says one row per item for this single check. Apply that policy rather than deleting every repeated value.
• A summary combines recorded values for a purpose. A total count adds quantities for different items in the same unit. It should not add category codes or mix incompatible units. The worked data is complete and fictional; the answer describes only these rows.

Useful vocabulary

field

One named piece of information in a record.

codebook

A guide to allowed recorded values.

identifier

A label used to distinguish items.

Watch your reasoning

• A field can be easy to collect but irrelevant to the purpose.
• Standardizing a known spelling is different from inventing an unknown value.
• Equal counts in different items are not duplicate records.
• A total loses meaning if kilograms and item counts are combined.

Make a practice plan

1. Draw or tell one computing idea from each of the four lessons. Give one example of using each idea.
2. Use the example to explain one mistake to avoid.
3. Try the unit review before opening the explanations.
4. Try the new task on paper or say your answer. Then compare your steps with the example answer.