Home / Statistical Tools / Control Charts / p Chart / Options
Options¶
Want to build one first?
How-To works through a finished chart step by step, from sample data you copy off the page straight into Excel. The data has a point that is out of control, so the result has something to read rather than twenty points in a row.
How-To with GroupBy does the same thing with a GroupBy column, so one run produces a separate chart per group.
This page covers the dialog behind all three P-family charts, because there is only one: the ribbon entry P, NP, and Laney's P' opens a single window whose Variant: radio row selects p, np or Laney P'. Everything here applies to all three unless it says otherwise, and the two variant pages, np Chart Options and Laney P' Chart Options, list only their differences.
The pane that opens beside the finished chart has its own page, Task Pane. What each option means statistically is on the Math Details page.
The Dialog¶
The dialog has three tabs: Data, Limits & Splits, and Rules & Display. Above them sits a Variant: row with P, NP and Laney P'; P is selected when the dialog opens. The window title follows the selection, so the same dialog reads P Chart, NP Chart or Laney P' Chart. The buttons along the bottom are Back, Help, Cancel and Finish.
There is no Capability tab, and no specification limit, target, capability index or histogram can reach an attribute chart's output. Also absent: the Transformation group and the Estimate Sigma group. An attribute chart's sigma is not estimated separately from its center line: the binomial model ties them together, so there is nothing to choose. Any historical sigma is ignored by the engine even if some other part of the product offers one.
Data Tab¶
The calculation type sits at the top as Excel and GroupBy, with a Data Types button beside it. Below that:
| Control | Type | Default | What it does |
|---|---|---|---|
| Defectives Column | slot list | empty | The count of defective items per subgroup. Checking several columns produces several charts, one per column |
| Sample Size | radio pair: Column or Constant | Constant | Where the number inspected comes from |
| n = | text box, live under Constant | 1000 | The constant sample size |
| X-Axis Labels (Optional) | slot list | empty | Labels drawn along the X axis |
| GroupBy | region with Available Items and GroupBy Order | empty | One chart per group |
Three things are worth knowing before you fill this in:
- The sample size default is a constant of 1000, not a column. If your sample sizes live in a column, switch to Column first; leaving it on Constant silently charts every subgroup as though 1000 items had been inspected.
- There is no subgroup group on this dialog. One row is one subgroup, and the sample size column says how many items that subgroup contained.
- Defectives cannot exceed the sample size. One offending row stops the analysis with a message naming the subgroup number. A negative count or a sample size of zero or less is refused the same way, and the chart needs at least two usable subgroups.
Limits & Splits Tab¶
Three groups: Control Limits, Baseline Estimation and Split Control Limits.
Control Limits¶
| Control | Type | Default | What it does |
|---|---|---|---|
| Shewhart / Manual / None | radio group | Shewhart | Whether limits are computed, typed, or omitted |
| Number of StDevs: | text box | 3 | The multiplier on the standard deviation of the plotted statistic |
| Historical pBar (optional): | text box | blank | Supplies \(\bar{p}\) outright instead of estimating it |
| Manual UCL:, Manual Center:, Manual LCL: | text boxes | blank | Used only under Manual |
| Force Straight Limits | check box | off | Substitutes one assumed sample size for the real ones |
| Assumed n: | text box | blank | The size used when Force Straight Limits is on |
Historical pBar must be a proportion. Supplied, it replaces the pooled estimate and the data no longer influences where the center line sits.
Force Straight Limits does nothing on a p chart. This is the one control on the tab that differs by variant, and the difference is not cosmetic:
| Variant | What an assumed n changes |
|---|---|
| P | Nothing. The engine never substitutes an assumed size on this variant, so the limits keep stepping with the real sample sizes |
| NP | The center line and the limits, both of which otherwise vary with the sample size |
| Laney P' | Nothing, for the same reason as P |
So on a p chart the check box and its box can be set and will be stored, and the chart will come out exactly as it would have without them.
Manual limits replace all three lines with the numbers you type, and they must be complete: a partial set is rejected for the whole chart. None draws the points and the center line with no limits, and with no limits there is nothing for the out-of-control tests to compare against.
Baseline Estimation¶
| Control | Type | Default | What it does |
|---|---|---|---|
| All Subgroups | radio button | selected | Estimates \(\bar{p}\) from every subgroup |
| First N Subgroups | radio button | not selected | Estimates from the first N only |
The count is a count of subgroups. Every subgroup is still plotted; only the estimate is restricted, and the resulting limits extend across the whole chart.
Split Control Limits¶
| Control | Type | Default | What it does |
|---|---|---|---|
| None | radio button | selected | One estimate for the whole chart |
| Column | radio button | not selected | A new phase begins wherever the chosen column's value changes |
| Manual Breaks | radio button | not selected | A new phase begins at row numbers you list |
| Split Column: | drop-down | empty | The column read under Column mode |
| Manual Break Rows (1-based, start of split): | text box + Add / Remove Selected | empty | The list of phase starting rows |
| Split Labels (shown beside each divider): | grid | empty | One editable label per phase |
Each phase gets its own \(\bar{p}\), its own limits and its own \(\sigma_Z\), and the summary table gains a Per-Split Breakdown block.
Rules & Display Tab¶
Out-of-Control Rules¶
Only Rule 1 is on when the dialog opens. The variable charts run all eight tests by default; these run one.
| Control | Default | Threshold box |
|---|---|---|
| Rule 1 - Beyond K sigma from center | on | none |
| Rule 2 - N consecutive points on same side of center | off | N = 9 |
| Rule 3 - N consecutive increasing or decreasing | off | N = 6 |
| Rule 4 - N consecutive alternating up/down | off | N = 14 |
Rules 5 through 8 are not on this dialog and are disabled in the engine, so the zone-reading tests cannot be run on an attribute chart at all. Rule 9, the zero-count Benneyan test, belongs to the G Chart and is not here either.
Each threshold box takes a whole number of 2 or more.
The chart does not tell you which rule fired
Every rule you tick is tested, and the results are combined into a single verdict per point: this point broke at least one enabled rule, or it did not. Points that broke one are drawn in a different color from the rest, and that is the whole of the report. There is no per-rule label, no per-rule marker, no per-rule column, and no list of violations anywhere in the chart or the worksheet. A point that breaks four rules at once looks exactly like a point that breaks one, and the output never names either.
So when you turn several rules on, read a marked point as an instruction to go and look at the chart, not as a diagnosis. If you need to know which pattern caused a particular signal, the practical method is to switch the other rules off and rebuild the chart.
Display Options¶
| Control | Type | Default | What it does |
|---|---|---|---|
| Show 1-sigma / 2-sigma zone lines | check box | off | Draws the four intermediate lines |
| Reference Lines: | grid with Value, Label, Color, Thickness, plus Add Reference Line and Apply | empty | Draws a horizontal line at a value you choose |
The zone lines on an attribute chart are one full standard deviation apart, not thirds of the drawn band. The engine takes the distance from the center line to the upper limit and divides it by the sigma multiplier, so at the default 3 the 1-sigma line really is one standard deviation out. Change the multiplier to 2 and the zone lines stay one and two sigma out while the limits move in to two. The lower zone lines are not floored at zero, so on a chart whose lower limit was clamped they can be drawn below zero.
A reference line on a p chart is drawn in proportions, so a target of 2% is entered as 0.02.
Diagnostics¶
This group exists only on the attribute dialogs.
| Control | Type | Default | What it does |
|---|---|---|---|
| Compute and report sigma_Z statistic | check box | on | Computes the overdispersion factor and writes it to the summary table |
| Include secondary Z-score plot | check box | off | Adds a plot of the standardized subgroup values |
\(\sigma_Z\) is reported on all three variants and applied by Laney P' alone. On a p chart it is advice; on a Laney P' chart it is the arithmetic. See Math Details.
Task Pane¶
The Control Chart pane that opens beside a finished chart has its own page: Task Pane.