Database Design Using Normalization Questions and Answers

Answer: B

Correct answer: The inconsistent values problem

Grouping on a column and counting the members of each group exposes values that should be the same but are not: variant spellings, different word orders and stray punctuation each form their own group. A quick look at the group list makes the inconsistency obvious in a way that scanning the rows would not.

Answer: A

Correct answer: first step.

Counting the rows and examining what columns exist is the first step, because it establishes the scale and the shape of what has arrived before any inference is attempted. Only then can functional dependencies and keys be worked out in the second step and the referential integrity constraints tested in the third.

Answer: B

Correct answer: False

Denormalized is precisely the opposite of BCNF: the tables have been recombined, or left combined, so that determinants which are not candidate keys remain. That is done deliberately to avoid joins, and it is a reasonable choice for read-only data, but it means accepting the duplication and the anomalies that BCNF would have removed.

Answer: A

Correct answer: True

Anomalies arising from functional dependencies come from determinants that are not candidate keys, and BCNF is defined precisely by the absence of those. Removing them removes the anomalies. Multivalued dependencies are a separate source and need fourth normal form, but for the functional case BCNF is sufficient.

Answer: B

Correct answer: commonly

Building a read-only database from a set of extracted tables is a well-defined, low-risk task with no update logic to get wrong, which is why it is a common first assignment. The design priorities are different too: since nothing will be updated, modification anomalies cannot arise and the tables are often deliberately left denormalized for speed.

Answer: A

Correct answer: True

A general-purpose remarks column collects whatever users think worth noting, so it holds different kinds of fact in different rows, in free text that cannot be validated or queried. It is one of the standard problems in received data, and the remedy is to identify what is actually being recorded there and give each item a column of its own.

Answer: B

Correct answer: False

There is no such SQL construct. The number and type of columns in a table are found by querying the system catalog or the information schema, or by using a product-specific command such as DESCRIBE or sp_columns. COUNT(*) does exist and counts rows, which is probably what the invented name is meant to echo.

Answer: B

Correct answer: False

Nothing about received data guarantees the absence of multivalued dependencies; the tables were designed by someone else, possibly with no normalization at all. Their presence must be established by examining the data and understanding what it means, and where one is found the relation should be split so that each holds a single theme.

Answer: B

Correct answer: False

The statement is right about the problem and wrong about its name. Missing values are called null values, not blank values, and the distinction matters because a null is not an empty string - it is the absence of a value, and it may mean unknown, not applicable, or simply not yet entered.

Answer: B

Correct answer: second step.

The second step is where the structure is inferred: the functional dependencies among the columns, and from them the primary keys, and then the foreign keys that link the tables. The first step is only a survey of what has arrived, and the third checks whether the implied referential integrity constraints are actually satisfied.