## Chapter 8 — Five sensitive is a Boolean sum ### Lesson map Lesson 08. Map, example, EX054. ```mermaid flowchart TD F["Five True flags"] --> S["sensitive 5"] F --> P["Two of those five still pass the filter"] P --> Q{"Is 5 a count of patients or of staff?"} Q -->|No| K["It is the Boolean sum"] ``` The flag is a classification bit. It is not a headcount. ### Logical drawing ```text True and kept: referrals/ at staff booking-app-config at admins True and bad: patient-letters/ , invoices/ , old-backup.zip False: rota.xlsx , price-list.pdf , logo.png ``` ### Learning outcomes After this chapter, you should be able to: 1. List the five objects whose flag is true. 2. Separate that sum from the failure list and from the staff count. 3. Refuse a kappa or a patient census read into the number 5. ### 8.1 The flag is Leila's label, counted once. The sensitive rule names treatment letters, referrals, invoices, backups, and the booking configuration. The script encodes that list as five `True` flags and three `False` flags. `sum` adds the flags. It does not open a letter, and it does not count patients. Two true flags still pass the membership test. `referrals/` is `staff`. `booking-app-config` is `admins`. Both strings sit in `allowed[True]`, so those triples increase `sensitive 5` and stay out of `bad`. A note that says "five sensitive items were shared too widely" has added two passing rows to a list that holds three. Eight is the staff count in the prose, and eight is also `len(items)`. Those eights name different sets: people, and top-level objects. The script has no person field. Landis and Koch (1977) define agreement between raters. One person applied these flags, no second coding is recorded, and no kappa is defined. The printed 5 is not an agreement statistic. ### 8.2 Worked example | Reading of the 5 | Ground | Decision | |---|---|---| | Five flags are true | The Boolean sum | Keep | | `referrals/` is sensitive and not a failure | `staff` is allowed | Keep | | "Five patients were exposed." | No patient field | Refuse | | "Two raters agreed, so kappa is high." | One coder, no kappa | Refuse | > **Synthetic case.** The flags encode the clinic's list. They do not count people. ### Errors and limits Do not borrow the staff total of eight as if it were the item total's meaning. ### Apply - **EX054 — What 5 counts.** A note says five sensitive items failed, that this equals five patients, and that two reviewers agreed. Which of those claims survive the Boolean sum? *Rubric:* 3 = only the flag count survives; referrals and booking-app-config are sensitive and allowed; the failure list stays three names; no patient count and no kappa;; 2 = 5 is flags, and two sensitive rows pass;; 1 = one claim;; 0 = none. ### Continue studying — lesson 08 (chapter 08 of 20) [← Chapter 07](#ch-07) · [Course chapter index](../SUBJECTS.md) · [Chapter 09 →](#ch-09) Continue to the next authored lesson. ## Worked answers **EX054.** Five is the Boolean sum. `referrals/` and `booking-app-config` are sensitive and still pass. The failure list stays three names. There is no patient count and no kappa.