Tutorials › Calculus I › Detecting Inconsistent Function Data

Function Notation and Representations · Tutorial 30 of 1495

Detecting Inconsistent Function Data

You will learn to test whether input-output data can represent a function and how to handle conflicts responsibly.

Beginner 10 min read

What You'll Learn

  • Check whether every listed input has exactly one assigned output.
  • Distinguish a repeated input-output pair from an input assigned conflicting outputs.
  • Identify when incomplete data are not inconsistent.
  • Diagnose conflicting entries in tables and sets of points.
  • Explain why inconsistent data do not determine a unique repair.
  • Choose a correction that is supported by the data source or modeling context.

When Do Input-Output Data Describe a Function?

A table or collection of points can look orderly and still fail to describe a function. The key question is not whether the outputs follow a familiar pattern. It is whether each input is assigned exactly one output.

For example, a table may list the same input twice. That is not automatically a problem: the repeated entries might give the same output. But if the same input is paired with two different outputs, the data do not describe a function of that input.

Definition: Let \(D\) be the set of inputs represented in a data set. The data describe a function on \(D\) if every input \(x\in D\) is assigned exactly one output \(y\). Equivalently, each input appears in at least one input-output pair, and any two pairs with the same input have the same output.

In a table, an input-output pair is a row \((x,y)\). In a collection of points, it is a point with horizontal coordinate \(x\) and vertical coordinate \(y\). The data are inconsistent as function data precisely when there are two listed pairs \((a,b)\) and \((a,c)\) with \(b\ne c\).

The word exactly matters. Every input in the stated domain must have an output, and it cannot have two different outputs. A data set may be incomplete because it gives no output for some input that is not listed. That does not make the listed data inconsistent; it means only that the data do not provide a value there.

A Direct Test for Conflicting Outputs

To check a finite table or point collection, group the entries by input. Then compare the outputs within each group. A group containing one distinct output is consistent. A group containing two or more distinct outputs is a conflict.

1
List the inputs.
Look for every input that occurs more than once.
2
Compare outputs for each repeated input.
Identical outputs are consistent repetitions; different outputs are conflicts.
3
State the conclusion precisely.
If there is a conflict, identify the input and the different outputs. If there is none, the listed pairs pass this test.

This test concerns whether the data can represent a function from input to output. It does not test whether the data fit a particular formula, are complete, or follow a pattern such as linearity. Those are separate questions.

Why the Test Works

The definition of a function requires a unique output for each input. If two entries give different outputs for the same input, uniqueness fails. Conversely, if each input appearing in the data has one distinct output, we can define a function on the set of listed inputs by assigning that output to each input.

Justification: Suppose the entries contain no input with two different outputs. Define \(f(x)\) to be the output paired with \(x\). Each listed input has an output by the way the data are recorded, and the no-conflict condition ensures that this output is unambiguous. Therefore \(f\) is a function on the set of listed inputs. If some input \(a\) has different outputs \(b\) and \(c\), then a function would have to satisfy both \(f(a)=b\) and \(f(a)=c\), which is impossible when \(b\ne c\).

This argument also explains why a repeated pair causes no trouble: repeating \(f(a)=b\) does not give \(a\) a second, different value.

Worked Examples

Worked Example: Checking a Table with a Repeated Pair

Determine whether the following table can describe a function from its first column to its second column.

$$ \begin{array}{c|c} x&y\\ 0&2\\ 1&4\\ 1&4\\ 2&6 \end{array} $$

The input \(0\) appears once and is paired with \(2\). The input \(1\) appears twice, but both entries pair it with \(4\). The input \(2\) appears once and is paired with \(6\). Thus no input has two different outputs.

The set of listed inputs is \(D=\{0,1,2\}\). We can define \(f(0)=2\), \(f(1)=4\), and \(f(2)=6\). Each input has exactly one output, so the table is consistent function data. The repeated row \((1,4)\) is redundant, but it does not create a conflict.

Check: Grouping the rows by input gives \(0\mapsto2\), \(1\mapsto4\), and \(2\mapsto6\). Every group has one distinct output, as required.

Worked Example: Finding a Conflict in a Point Collection

Decide whether the points \(\{(1,3),(2,5),(1,4),(3,7)\}\) can be the graph of a function \(y=f(x)\).

Read each point as an input-output pair. The input \(1\) occurs in \((1,3)\) and \((1,4)\). Its listed outputs are \(3\) and \(4\), and \(3\ne4\). The input \(2\) has the single listed output \(5\), while input \(3\) has the single listed output \(7\).

The two points with first coordinate \(1\) mean that the proposed data require both \(f(1)=3\) and \(f(1)=4\). A function cannot satisfy both equations. Therefore the collection is not the graph of a function of \(x\).

Check: The conflict is located specifically at input \(1\), with outputs \(3\) and \(4\). Removing either of those two points would eliminate this particular conflict; leaving both points in the collection would not.

Worked Example: Finding and Reporting a Table Conflict

Check the table below, then state what can and cannot be concluded about correcting it.

$$ \begin{array}{c|c} x&y\\ -1&0\\ 0&2\\ 2&6\\ 2&7\\ 4&10 \end{array} $$

The inputs \(-1\), \(0\), and \(4\) each appear once, so each has one listed output. The input \(2\) appears twice, paired with \(6\) and \(7\). Since \(6\ne7\), the table does not describe a function on its listed inputs.

The conflict can be reported as: “At \(x=2\), the table gives both \(y=6\) and \(y=7\).” The data alone do not tell us which entry is wrong. If the first \(y\)-value is correct, changing the second entry from \(7\) to \(6\) would remove the conflict. If the second is correct, changing the first from \(6\) to \(7\) would also remove it. If an original record shows that one value was copied incorrectly, that source can justify a particular correction.

Check: After either proposed single-entry change, input \(2\) has only one distinct output, and all other listed inputs already have one. Without additional information, however, we cannot claim which proposed change recovers the intended data.

How to Repair Inconsistent Data Responsibly

A conflict shows that something about the data or their interpretation must be clarified before they can represent a function of the stated input. It does not, by itself, reveal the correct value. A responsible repair identifies the conflict first and then uses reliable information to decide what to change.

  • Check for a recording or transcription error. Compare the conflicting entries with the original measurement or record. Correct a value only when the source supports the correction.
  • Check whether the input was recorded precisely enough. Two entries that look identical may have come from different underlying inputs that were rounded or grouped. If so, the input description may need greater precision.
  • Check whether the input needs another variable. If the output genuinely depends on an additional condition that was omitted, the stated input may not contain enough information to determine one output. Add the relevant condition only if it belongs in the model.
  • Remove an entry only with a reason. A duplicate record, invalid observation, or known data-entry error may justify excluding an entry. Deleting a value simply because it creates a conflict does not establish that the remaining data are correct.

Averaging two conflicting outputs is not an automatic repair. For example, replacing \(6\) and \(7\) by \(6.5\) invents a new output; it does not show that \(6.5\) was the recorded or intended value. An average may be appropriate in a context with a stated averaging rule, but that rule is additional information, not a consequence of the function definition.

Key takeaway: Detecting a conflict is a mathematical conclusion; choosing a correction generally requires evidence or a stated modeling rule. Do not present one possible repair as the uniquely correct repair unless the available information supports it.

Related Checks and Common Misreadings

A frequent mistake is to treat every repeated input as a conflict. Consider the entries \((5,8)\) and \((5,8)\). Both assign the same output to input \(5\), so they are consistent. The conflict test compares distinct outputs, not the number of times an input is written.

Another mistake is to conclude that data are inconsistent merely because they do not fit a familiar formula. For example, the pairs \((0,1)\), \((1,3)\), and \((2,6)\) each give a different input one output. They are function data, even if a proposed rule does not match them. A conflict would require the same input to have different outputs.

For point data on a graph, the same condition can be seen geometrically: two points with the same horizontal coordinate but different vertical coordinates lie on the same vertical line. That line meets the collection at two different points, so the collection cannot be the graph of \(y=f(x)\). The algebraic check and the graphical check express the same uniqueness requirement.

Finally, a missing row is not automatically a conflict. If the listed inputs are \(0\) and \(2\), the table simply gives no listed output for \(1\). If the intended domain is only \(\{0,2\}\), the entries can define a function on that domain. If the intended domain also includes \(1\), the table is incomplete for that domain, but incompleteness is different from assigning two outputs to one input.

Check Your Understanding

For each question, focus on which inputs occur and how many distinct outputs are paired with them.

  1. Do the pairs \((2,5),(2,5),(3,7)\) describe a function on their listed inputs? Explain why.
  2. Which input creates a conflict in \((0,1),(1,2),(0,4),(2,5)\), and what are its conflicting outputs?
  3. If a table has one row for \(x=1\) and no row for \(x=2\), is that necessarily inconsistent? Distinguish the answer from whether the table is complete on a domain containing both inputs.
  4. A point collection contains \((4,6)\) and \((4,9)\). Explain why it cannot be the graph of \(y=f(x)\).
  5. Two entries assign outputs \(10\) and \(12\) to the same input. What evidence could justify a repair, and why is replacing them by \(11\) not automatically valid?

Test yourself: Calculus I final exam practice

Ready to check what you know? Work a full-length practice final, then open the complete worked solutions.

Want all 1,114 practice problems?

This tutorial covers one topic. Calc Pass covers the full Calculus AB & BC curriculum (the core of Calculus I and II): full review, every practice question, and step-by-step solutions to each one.

QR code to download the Calc Pass app

Get the Calc Pass App

Scan to download and practice on the go, with instant scoring.

Calc Pass Volume 1: 1,114 Practice Questions book cover

Vol. 1: 1,114 Questions

The complete practice question bank, organized by AB & BC topic.

View on Amazon
Calc Pass Volume 2: Solutions Manual book cover

Vol. 2: Solutions Manual

Fully worked solutions to every question in Volume 1.

View on Amazon