A WORKSHOP ON COMMUNICATING YOUR WORK

Good work does not speak for itself

Communication is where the work gets finished.

This is for anyone whose work ends in a decision someone else makes: specialists, policy and product staff, researchers and school leaders writing for the people above them. It deals with papers, briefs, decks, evaluations and think pieces.

About the examples. The cases in this session come from practice and are anonymised. Names, organisations and details have been removed, changed or combined, and some cases are invented to make a point, so treat any example as potentially fictional. The research is real, and every study cited is in the references.

PREDICT BEFORE YOU LOOK · ILLUSTRATIVE

Two openings. Same evidence. Same author.

Which one gets endorsed? Which one took longer to write?

VERSION A

"Background."

• In March the team began a review of teacher use of the assessment module.
• We interviewed 24 teachers in 8 schools and analysed six months of usage logs.
• Several themes emerged and are discussed below.
• Section 5 sets out recommendations for consideration.

VERSION B

"Decision sought: approve a redesign of the marking flow before the Term 3 release."

• Teachers abandon the rubric step more than any other, and most do not come back that day.
• A redesign tested with 24 teachers cut abandonment in testing.
• We are fairly confident it will hold across schools and less sure about very large classes.
• Detail and risks follow.

Both openings were written for this session from the same invented case. Commit to two answers before reading on: which version a busy decider is more likely to endorse, and which one cost its author more time.

Most people pick B for the first question and hesitate on the second. B took longer, and the reason it took longer is the subject of everything that follows.

THE REVEAL

Only one of them is finished.

A is a record of the work. B is the work, done.

B took longer because it forced decisions that A could postpone: what the reader must do, what the claim is and how sure the author is. A lets the author stop at the evidence and hand the reader the job of concluding.

That postponement is what this session is about. The pair returns at the close.

Part one of five

Why work fails to travel

The problem, and why it is a thinking problem.

• ONE Why work fails to travel
• TWO Decide what to say
• THREE Earn belief
• FOUR Make it land
• FIVE Make it yours

The first part names the problem and argues that it is a thinking problem before it is a writing problem.

01

DIAGNOSE

Part one of five

01

Most work is told the way it was done

We report in the order we found things. Readers need the order they will use them in.

Most papers, briefs and decks that fail to move a decision do not fail for lack of evidence. They fail because the evidence arrives in the order the author met it, with the point at the end and nothing to tell the reader how much weight it carries.

This section names the default, explains why it is the default and separates the job here from a neighbouring one: telling people about a change that has already been decided.

THE DEFAULT

We report in the order we found things. Readers need the order they will use them in.

Writer-based prose is the default, not a lapse.

Flower (1979) named it writer-based prose: organised as a narrative of the writer's discovery, built around the writer's own code words and surveying the topic instead of making a point. Reader-based prose is rebuilt around what the reader needs to know and do.

Most first drafts are writer-based, and that is fine. Drafting is how the writer finds out what they think. Shipping the draft is the problem.

A quarterly update opened: "In January the team met the vendor. In February we ran workshops." The decision it needed, whether to renew the contract, sat on page four. Rewritten, page one said: renew for one year, on two conditions.

TWO WAYS TO WRITE

Telling what you know is not the same as building what they need

KNOWLEDGE TELLING

Retrieve and report

• Content arrives in the order it comes to mind
• Report the sequence of the work
• Done when everything is in

KNOWLEDGE TRANSFORMING

Decide and rebuild

• Decide the goal first
• Select and reorganise to serve it
• Done when the reader can act

Bereiter and Scardamalia (1987) contrast knowledge telling, where writers retrieve content in the order it comes to mind, with knowledge transforming, where the writing goal reshapes the content itself. Expert writers transform.

Transforming is harder because it changes what the writer thinks, not only how it is said. That is why it cannot be left to an editor at the end.

THE SYMPTOMS

Six signs work will not travel

• 01 The reader has to dig for the point
• 02 Every finding gets equal weight
• 03 Claims arrive with no confidence attached
• 04 Polish covers thin thinking
• 05 One document tries to serve every reader
• 06 Nobody checked what the reader understood

01The point is buried The claim sits in the last section or is never stated. The reader has to assemble it, and different readers assemble different claims.

02Everything weighs the same Findings appear in the order they were found, each with a paragraph of similar length. The reader cannot tell the finding that matters from the one that was merely interesting.

03No confidence attached A replicated result and a hunch arrive in the same voice. The reader either trusts both or neither.

04Polish over thinking Good design and confident prose make a thin argument look finished. The gap shows when someone asks the second question.

05One document for everyone A paper written for the minister, the vendor and the school at once serves none of them well.

06No check on what was heard The work is sent and assumed to be understood. Nobody asks the reader to say what they took from it.

Each sign maps to a later section. The first two are framing and structure problems, the third is calibration, the fourth is construction, the fifth is shape and the sixth is landing.

THE CLAIM OF THIS SESSION

Communication is not the wrapping on the work. It is where the work gets finished.

If the claim cannot be written before the paragraphs beneath it, the analysis has not reached a conclusion.

People overrate how well they understand something until they try to explain it (Rozenblit & Keil, 2002). Students who built slides with a claim as each headline understood the material better themselves than students who built topic slides (Aippersbach et al., 2013).

Writing the claim is a test of the analysis, not a summary of it. When the sentence will not come, the problem is upstream of the words.

SCOPE

This is not change communication

Telling people what changed is a different job from persuading people what is true.

CHANGE COMMUNICATION

A decision already taken

• Many readers, mostly users
• Travels down a cascade
• Asks people to adapt or prepare
• Fails by loss at each step

WORK COMMUNICATION

An argument not yet accepted

• Few readers, mostly people who can say no
• Travels up and sideways
• Asks people to decide, endorse or build on it
• Fails by burying the claim or overclaiming

Craft guide 20, Communicating change to schools, covers the cascade, the channel and the calendar. This session covers the other case: a few readers who can say no, judging an argument they have not yet accepted.

The two share some tools, such as leading with what the reader must do. They differ in what success looks like. Change communication succeeds when people know what to do. Work communication succeeds when the right person decides well.

THE TEST

The reader is the user.

Judge the work by what the reader can do after it, not by what you put in it.

This is the move UX made for interfaces: stop measuring output and start measuring what the person can now do. The curse of knowledge explains why the sender cannot see the gap from inside it (Camerer et al., 1989). Hinds (1999) found that experts were worse than intermediates at predicting how long novices would take on a task.

The full treatment of the curse of knowledge lives in craft guide 20 and the cognitive biases field guide. Here it is enough to know that the author is the worst-placed person to judge whether the work is clear.

An analyst sent a forty-page evaluation and counted it done. Asked a week later what the report recommended, the director named the one finding from the summary that the analyst thought least important.

Part two of five

Decide what to say

Who it is for, what the claim is and how it is organised.

• ONE Why work fails to travel
• TWO Decide what to say
• THREE Earn belief
• FOUR Make it land
• FIVE Make it yours

The second part covers the decisions that come before any sentence: who the work is for, what it claims and how the argument is organised.

02

AIM

Part two of five

02

Start from the reader's decision, not your content

Name the reader, the act and the date before you write a word.

Every piece of work that asks for something has a reader who can give it. Writing starts there: who that person is, what they can do with the work and by when they have to do it.

This section covers naming the act, telling decision work from work that changes how people think and mapping who can act, advise or block.

THE FIRST MOVE

Name the reader, the act and the date.

Write the act as a verb: approve, fund, defer, direct, decline.

A brief that cannot name the act is a paper for the file, and should be called one. The rhetorical situation has three parts: a problem that demands a response, an audience able to act on it and the constraints on what can be said (Bitzer, 1968).

Patton's (2008) test for evaluation is the same test in other words: intended use by intended users. If you cannot say who will use the work and for what, you cannot judge whether it is finished.

"For the deputy director: approve or defer the extension of the reading pilot, before the budget meeting on 30 June." One line, and every later paragraph has a job: help that person make that call by that date.

TWO KINDS OF WORK

Some work asks for a decision. Some changes how people think. Say which.

DECISION WORK

One act, one date

• Options, including doing nothing
• A recommendation
• A deadline that matters

ENLIGHTENMENT WORK

A reframing

• A new way to see the problem
• A horizon, not a deadline
• A date to revisit it

Weiss (1979) found that research rarely feeds a single decision. More often it shifts how people frame a problem over time, which she called enlightenment. Both are legitimate uses of evidence.

The failure is writing enlightenment work as if it asked for a decision, or the reverse. Think pieces are usually the second kind, and they read better when they say so.

A paper on attention and devices went to a committee marked "for decision". The committee looked for an ask, found none and deferred it. Reissued as a reframing with a date to revisit, it was discussed for an hour.

THE READERS

Map who can act, advise or block

Write for the person with the decision in their gift, then for the people they will ask.

FIG 01 · WHO CAN ACT ON IT · ILLUSTRATIVE

Managers attend to stakeholders in proportion to their power, legitimacy and urgency (Mitchell et al., 1997). Read in reverse, the same map tells a writer which findings have a lever behind them and which do not.

The person who advises the decider is often the real first reader. If the adviser cannot restate the claim, the decider will hear a weaker version of it.

FROM PRACTICE

A change of audience changes the claims you are entitled to make.

A conference talk written inside a government agency was later cleared for a public audience. The identifiers came out, and the stance shifted from an official position to what educators hope to achieve. The evidence did not change. What the speaker could claim, and on whose behalf, did.

The same holds for any piece that moves between readers. Before reusing work, ask who the new reader is and what you are entitled to say to them.

AIM IN PRACTICE

Aim in practice

DO

Habits that travel

• Name one reader and their act
• Write a separate first line for each reader
• Say whether you want a decision or a rethink

DO NOT

Habits that stall

• Write "for information"
• Address "all stakeholders"
• Leave the reader to work out what you want

A learning-science primer scoped for one design team was later widened to every product partner. The examples, the vocabulary and the density all had to change. A guide series written for three readers, a policy team, a vendor and a school lead, gave each its own section instead of one compromise.

A budget paper opened with one line for each reader. For the director: approve or defer. For the finance adviser: the cost, and the risk if it slips. For the school leads: what changes for them and when. Each reader found their line in ten seconds.

03

FRAME

Part two of five

03

Say the point before you prove it

If the claim cannot be written first, the analysis is not finished.

Analysis runs from evidence up to a claim. Reading runs the other way. The cost of communication is the cost of turning the work over so the claim comes first.

This section covers the inversion, five ways to open and the difference between a thesis and a slogan.

THE TEST OF THE ANALYSIS

If you cannot write the claim first, the analysis is not finished.

When the claim will not come, go back to the analysis, not to the wording.

Minto's (2009) rule is the answer at the top, then the grouped reasons beneath it, each answering the question the line above it raises. The structure is easy to describe and hard to fill, because it demands a conclusion.

A draft that opens with background is often a sign that the conclusion has not been reached yet. The fix is more thinking, not better prose.

An analyst could not write the claim for a survey report. The trouble was not the prose. Two findings pointed in opposite directions and nobody had decided which mattered more. Once that was settled, the opening took ten minutes.

THE INVERSION

Analysis climbs. Delivery descends.

You build from evidence up to a claim. The reader needs the claim first and the evidence beneath it.

FIG 02 · BUILT UPWARD, READ DOWNWARD

This inversion is why communication costs effort. The order you thought in is the wrong order to be read in, so the work has to be rebuilt, not reformatted (Minto, 2009).

Every move in the rest of this session is a way of paying for the inversion: naming the reader, writing the claim, showing how it was built and saying how sure you are.

OPENINGS

Five ways to open

• BLUF Bottom line up front
• SCQA Situation, complication, question, answer
• INTENT Purpose, key tasks, end state
• CONTRAST Not X, but Y
• CARS Territory, gap, answer

BLUFBottom line up front The recommendation, conclusion or reason for writing goes in the first or second paragraph, not at the end (Department of the Army, 2020). The default for decision work.

SCQASituation, complication, question, answer Set out what the reader already accepts, what has changed, the question that raises and your answer to it (Minto, 2009). Useful when the reader needs a moment of context before the answer lands.

INTENTCommander's intent State the purpose, the key tasks and the end state, so people can act sensibly when the plan breaks (Department of the Army, 2019). Useful for work that hands a task to others.

CONTRASTThe contrast line State what everyone assumes, then what is actually true. A strong contrast can carry a whole paper, as long as it can be unpacked.

CARSCreate a research space For work that has to earn attention before it can ask for anything: establish the territory, show the gap and occupy it with your answer (Swales, 1990). The default for enlightenment work.

Match the opening to the work. BLUF suits decision work, where the reader needs the ask. CARS suits enlightenment work, where the reader first needs a reason to care.

IN ACTION · ILLUSTRATIVE

One finding, three openings

Same evidence. The opening depends on what the reader needs first.

BLUF

For a decision

• "Approve the marking-flow redesign before Term 3. Teachers abandon the rubric step more than any other."

SCQA

For a reader who needs context

• "Teachers mark in the module every week. Since March most leave at the rubric step. Can we keep them in? A shorter rubric did, in testing."

CARS

For a reader who needs a reason to care

• "Marking tools are judged on features. Nobody measures where teachers give up. We did, and it is one step."

All three openings are invented, from the marking-flow case used throughout. Each puts the claim in the first two sentences. They differ in what comes before it, and that depends on what the reader already accepts.

BLUF assumes the reader knows the problem. SCQA spends one sentence on shared ground first. CARS earns attention before it asks for anything.

A CONTRAST LINE THAT CARRIES A PAPER

From a think piece on AI and growing up.

The aim is not simply to prepare students to use increasingly capable AI, but to ensure growing up with AI leaves them increasingly capable as humans.

A think piece on AI and growing up · anonymised

The line works because it reverses the default framing, and every section of the paper can be traced back to it. Protect, cultivate, redesign and study each answer the question it raises.

Try it on your own work before reading on. Write your claim in 25 words. Then ask someone else to say what you want them to do. If they cannot, the claim is not yet a claim.

THESIS OR SLOGAN

A thesis can be unpacked. A slogan cannot.

DO

Habits that travel

• Open with the claim
• Make the contrast do analytical work
• Trace every section back to the claim

DO NOT

Habits that stall

• Open with background
• Use a line that sounds good but cannot be unpacked
• Save the recommendation for the end

One talk on technology in schools rested on three parallel shifts. Less of the tool, more of the change in practice. Less customisation, more agency. Less scaling fast, more compounding. Each one opened a section, and each could be argued.

The test is tracing. Anything that cannot be traced back to the claim is either a second paper or decoration.

04

STRUCTURE

Part two of five

04

Make the logic visible at a glance

Headings should carry the argument even if nobody reads the body.

Most readers read the headings and the first lines, then decide whether to go on. If the headings name topics, that reader learns the table of contents. If they state claims, that reader learns the argument.

This section covers claim headings, the ghost deck, ordering by logic and the rules for grouping.

HEADINGS

Headings should carry the argument even if nobody reads the body.

Write headings as claims, not topics.

Garner and Alley (2013) tested this on slides. 110 engineering students saw either sentence headlines supported by visual evidence or topic headings with bullets. The headline group showed better comprehension, fewer misconceptions and lower reported load.

The same discipline applies to report sections, email subjects and slide titles.

A steering committee pack had the headings Background, Approach, Findings and Next steps. Rewritten, they read: Teachers skip the rubric step. Shorter rubrics keep them in. Redesign before Term 3. A member who read only the headings could now vote.

LABEL OR ARGUE

Topic headings label. Claim headings argue.

TOPIC

Tells you what is coming

• Teacher trust
• Findings
• Next steps

CLAIM

Tells you what to believe

• Teachers trust tools they can predict
• Abandonment peaks at the rubric step
• Redesign before Term 3

A topic heading tells the reader what is coming. A claim heading tells them what to believe and lets them decide whether to read on for the proof (Garner & Alley, 2013).

Claim headings also expose weak sections. A section whose heading cannot be written as a claim usually has nothing to claim.

THE GHOST DECK

Read only the headings. Does the argument survive?

Write every heading first, read the column top to bottom, then fill in the evidence.

FIG 03 · THE SAME PAPER, READ BY ITS HEADINGS

The ghost deck writes every heading first and reads the column top to bottom before any body text exists. If the column does not argue, no amount of body text will.

Reverse outlining runs the same test on a finished draft. In one margin, write what each paragraph says. In the other, write how it moves the argument, in 5 to 10 words each. A paragraph that cannot be summarised that briefly needs rewriting (Purdue Online Writing Lab, n.d.-d).

ORDER

Order by logic, not by chronology

CHRONOLOGY

The order of the work

• What we did first
• What we did next
• What we found

LOGIC

The order of the argument

• Why it matters
• What to do
• How to do it
• Whether it works

One paper was organised as Protect, Cultivate, Redesign and Study. A series of practice guides was grouped by the life of a product: why build, what to build, how to build, data and AI, and whether it works.

Parallel phrasing tells the reader which ideas are peers. Chronology tells them only what the author did on which day.

An evaluation report ran in the order of the project: set-up, term one, term two, findings. Rebuilt in the order of the argument, it opened with what worked, then what to change, then what it would cost and how anyone would know. Nothing was added. Every paragraph moved.

GROUPING

Grouping rules

• 01 No overlaps, no gaps
• 02 Same kind of idea in each group
• 03 Old before new
• 04 One idea per paragraph
• 05 Parallel ideas, parallel words

01No overlaps, no gaps Groups should not overlap and together should leave nothing out (Minto, 2009). If an item fits two groups, the grouping is wrong.

02Same kind of idea Ideas in a group are the same kind of idea: all reasons, all steps or all problems. A list that mixes them hides its own logic.

03Old before new Put known information at the start of a sentence and the new point at the end, where readers expect emphasis (Gopen & Swan, 1990).

04One idea per paragraph State the idea in the first sentence. The rest of the paragraph supports it.

05Parallel words Ideas that are peers should be phrased as peers. Parallel phrasing is a signal the reader uses without noticing.

The five rules apply at every scale, from the sections of a paper to the items in a single list.

Part three of five

Earn belief

How the claim was built and how sure you are.

• ONE Why work fails to travel
• TWO Decide what to say
• THREE Earn belief
• FOUR Make it land
• FIVE Make it yours

The third part is about earning belief: showing how a claim was built and saying how sure you are.

05

CONSTRUCT

Part three of five

05

Show how the claim was built

Readers trust claims they can see being built.

A claim earns belief when a sceptical reader can trace it down to what was recorded and see the step that links the two. Most workplace writing states the claim and the grounds and leaves the link unspoken.

This section uses three old tools to make the construction visible: Gowin's Vee, stasis theory and Toulmin's layout of an argument.

BEFORE YOU WRITE

Fill the Vee before you write

What did you ask, what did you record, what do you claim and what do you value?

FIG 04 · ADAPTED FROM NOVAK & GOWIN, 1984

Gowin's Vee was built to make the structure of knowledge-making visible (Novak & Gowin, 1984). The focus question sits at the top. The thinking side holds world view, theories, principles and concepts. The doing side holds records, transformations, knowledge claims and value claims. Events and objects sit at the point.

Here it is adapted as a pre-writing check. Every claim on the right should trace down to a record.

Appendix A fills the Vee for the invented marking-flow case used throughout. Its value claim, that teacher time matters more than feature breadth, is the line most readers would argue with, which is why it is written down.

TWO KINDS OF CLAIM

Knowledge claims and value claims are settled differently

KNOWLEDGE CLAIM

"Abandonment peaks at the rubric step."

• Settled with evidence
• Can be checked
• Disagreement is resolvable

VALUE CLAIM

"Teacher time matters more than feature breadth."

• Settled by deliberation
• Is a choice
• Disagreement is legitimate

My rule for disagreement is to ask first whether it is about evidence, which can be resolved, or about values, which is a choice. The Vee keeps the two apart on the page (Novak & Gowin, 1984).

A paper that blends them invites the wrong argument. People attack the evidence when what they object to is the values.

STASIS

Find where the disagreement sits

Fact, definition, quality or policy. Most papers argue policy to readers still stuck at fact.

FIG 05 · THE STASES, IN ORDER

Stasis theory, from Hermagoras and later refined by Cicero and Quintilian, asks four questions in order: what happened, what kind of thing it is, how serious or how good it is and what to do (Purdue Online Writing Lab, n.d.-f). A policy argument cannot land while the reader still disputes the facts or the definition.

Brizee (2008) adapted the stases for workplace teams, who can agree level by level and act together while still disagreeing at one level. This sharpens the evidence or values rule with the case most workplace disputes turn on: definition. Is this adoption, or just log-ins?

THE PARTS OF AN ARGUMENT

Lay out the argument

Claim, grounds, warrant, qualifier, rebuttal. Most arguments at work state two of the five.

FIG 06 · A ROLLOUT ARGUMENT, LAID OUT

Toulmin (1958) separated the parts an argument needs: the claim, the grounds for it, the warrant that links grounds to claim, the backing for the warrant, the qualifier that states its strength and the rebuttal that names when it fails.

Most workplace arguments state grounds and claim and leave the rest unspoken. The layout makes the missing parts visible.

THE WEAKEST LINK

Underline the warrant. It is the part that has to travel.

Name the step from grounds to claim, and you have named where it can break.

The warrant is usually unstated because the writer finds it obvious. It is also where most claims break when they move to a new context: what held in volunteer schools may not hold in the rest.

The same layout works for appraising claims made by others, such as a vendor's evidence of impact. The next slide does exactly that.

IN ACTION · ILLUSTRATIVE

A vendor's claim, laid out

The pitch states grounds and a claim. The layout finds the warrant and where it breaks.

AS PITCHED

"Schools using our platform raised maths scores by 12 per cent."

• Grounds and a claim, nothing else
• Sounds like evidence of impact

LAID OUT

Grounds, warrant, rebuttal

• Grounds: 40 schools that chose the platform raised scores
• Warrant, unsaid: what worked where schools chose it will work where they did not
• Rebuttal: unless the schools that chose it were already improving

The vendor and the figure are invented. The pattern is common. A result from schools that opted in is offered as evidence for schools that did not, and the step between the two is never stated.

Laid out, the question to ask is obvious: how did comparable schools without the platform do over the same period? If the vendor cannot answer, the claim is still a hypothesis.

FROM PRACTICE

Removing a name does not remove the debt.

Source discipline changes conclusions, not just citations.

One research brief was first built on a colleague's unpublished talk. The first instinct was to drop the colleague's name and keep the findings. That fails, because the data and the quotes still belonged to someone else.

The brief was rebuilt on published evidence only, and two conclusions changed. "Negative scenarios work better" became "hard scenes cut both ways".

UNPACKING

Five questions to unpack any piece of work

• 01 What is the telling question?
• 02 What are the key concepts?
• 03 How was it investigated?
• 04 What is claimed as known?
• 05 What is claimed as worth doing?

01The telling question What question is the work trying to answer? If it cannot be stated, the work has no focus.

02The key concepts Which ideas does the work rely on, and does the reader share them?

03The method How were the records made and transformed? A reader cannot weigh a claim without knowing how it was reached.

04The knowledge claims What does the work claim to know? Each should trace to a record.

05The value claims What does the work claim is worth doing? These are choices and should be argued as choices.

These are Gowin's (1981) questions for unpacking a knowledge claim, usable on your own draft or on someone else's. The exact wording of the five is paraphrased here and listed for checking in the verification register.

Add one more: verify every source before it ships. A verification pass on one published guide found a fabricated author name and a paper cited for a finding it does not report. AI assistance makes this step more necessary, not less.

06

CALIBRATE

Part three of five

06

Say how sure you are, in a form people can use

A hedge is not a confidence level.

Readers have to act on claims of uneven strength. They can only do that well if the writer says which claims are solid and which are guesses, in terms the reader will read the same way.

This section covers evidence bins, the trouble with probability words and what the evidence says about stating uncertainty openly.

THREE BINS

Sort every claim into one of three bins

ESTABLISHED

Replicated and reviewed

• Holds across settings
• Safe to build on

EMERGING

Early and promising

• Limited samples
• Watch and test

EXTRAPOLATED

Reasoned from elsewhere

• Untested here
• Treat as a bet

One think piece on AI and growing up separated established science, emerging evidence and untested extrapolation, and said in the text which was which. Readers could then disagree with the bets without discarding the science.

The same logic appears as certainty ratings for evidence in health and as security ratings in education evidence toolkits.

PREDICT BEFORE YOU LOOK

"Likely" means different numbers in different heads

Write down the number you mean by "likely" before you look.

FIG 07 · THE SPREAD IS ILLUSTRATIVE, NOT ONE STUDY'S DATA

Kent (1964) found that intelligence analysts read the same estimative words as very different probabilities. Mauboussin and Mauboussin (2018) found the same spread in a large public survey.

The IPCC answered by fixing words to ranges: "likely" means 66 to 100 per cent and "very likely" 90 to 100 (Mastrandrea et al., 2010). A shared scale turns a word back into information.

A risk register called a supplier delay "likely". The project lead read it as about 60 per cent. The finance lead read it as near certain and froze a payment. Writing "about 60 per cent" would have cost three words.

WHAT THE EVIDENCE SAYS

Stating uncertainty in numbers barely dents trust. Vague caveats cost more.

5,780

Participants across five experiments, including a field test on a news website

Range

Numeric ranges: a small fall in trust

Words

Verbal caveats: a larger fall

Five experiments, including a field test on the BBC News website, with 5,780 participants in total. Stating uncertainty as a numeric range produced only a small fall in trust in the number and in its source. Verbal statements of uncertainty did more damage (van der Bles et al., 2020).

The fear that admitting uncertainty will cost credibility is mostly misplaced. Vagueness costs more than honesty.

THE RULE

Quantify, don't hedge.

Say what you are sure of, how sure and what would change your mind.

My rule is to qualify only where the qualification changes the meaning. A reflexive "may potentially" lowers trust without adding information. A range, a tier or a stated condition raises it (van der Bles et al., 2020).

The last clause is Cromwell's rule: leave room to be wrong and name what would show it. The Working Defaults field guide treats it in full.

IN ACTION · ILLUSTRATIVE

One finding, hedged and calibrated

HEDGED

"The programme may potentially have had some positive impact on attendance."

• Every word a hedge
• Nothing a reader can act on

CALIBRATED

"Attendance rose 3 to 5 points in the six pilot schools."

• Emerging: one term, no comparison group
• A second term with a comparison would settle it

Both versions are invented and report the same data. The hedged one sounds careful but tells the reader nothing about size, scope or strength. The calibrated one gives the range, the scope, the bin and what would change the author's mind.

The second version admits more doubt than the first, and it is the one a reader can trust (van der Bles et al., 2020).

THE COMMONEST OVERCLAIM

A small, conditional finding sold as a large, general rule.

Most claims are not false. They are inflated.

One review of a catalogue of education myths reached an unexpected conclusion: almost nothing in it was simply false. Most entries were small, conditional findings inflated into general rules.

The same inflation happens in our own briefs, one adjective at a time. "Improved" becomes "transformed" and "in two schools" becomes "in schools".

HABITS

Calibration habits

• 01 Numbers over adjectives
• 02 Show the counter-evidence
• 03 Date the evidence
• 04 Print the open questions
• 05 Name what would change your mind

01Numbers over adjectives Use effect sizes, ranges and counts. Adjectives inflate.

02Counter-evidence Where a finding is contested, show the other side in the text. One design brief carries it as its own marked paragraph beside each claim.

03Date the evidence Evidence ages. Say when a claim was last checked, as craft guide 16 recommends.

04Open questions Print what is still unverified, in the open. This document ends on its own verification register.

05What would change your mind Name the finding that would overturn the claim. It tells the reader how firmly you hold it.

Each habit costs a sentence. Together they let a reader act on the strong claims and hold the weak ones lightly.

Part four of five

Make it land

Words, surfaces and the room.

• ONE Why work fails to travel
• TWO Decide what to say
• THREE Earn belief
• FOUR Make it land
• FIVE Make it yours

The fourth part is about making the work land: the words, the surfaces and the room.

07

WRITE

Part four of five

07

Make every word earn its place

Plain words, short sentences, nothing the reader has to decode.

Most writing rules come down to four expectations readers bring to any text. Break one and the reader starts guessing at what you meant.

This section covers those expectations, the cost of long words, one worked rewrite and the difference between a principle and a preference.

THE FOUR MAXIMS

Four maxims under every writing rule

QUANTITY

Cut

• As much as needed, no more

QUALITY

State your confidence

• Only what you can support

RELATION

Lead with their question

• Stay relevant

MANNER

Plain words, clear order

• Be clear and orderly

Grice (1975) argued that readers assume a writer says as much as needed and no more, says only what they can support, stays relevant and is clear and orderly. Every rule in this section is one of those four.

When a writer breaks one, readers infer a hidden meaning. That is why padding reads as evasion. The Working Defaults field guide carries the maxims in full.

PLAIN WORDS

Long words do not make you sound clever. They make you sound less clever.

Plain language is not writing down. It is writing so senior readers can disagree with you quickly.

In five studies, needlessly complex vocabulary lowered readers' ratings of the author's intelligence (Oppenheimer, 2006).

Senior readers have the least time and the most to decide. Plain words respect both.

A REWRITE

Before and after

BEFORE · 25 WORDS

A wind-up

• "This paper seeks to explore the various ways whereby AI may potentially impact the developmental trajectories of learners, with a view to informing future considerations."

AFTER · 19 WORDS

A claim

• "AI is changing how children grow up. This paper sets out what schools should protect, cultivate, redesign and study."

Three things changed. The claim moved to the front. Nouns became verbs: "impact the developmental trajectories" became "changing how children grow up". The hedge went: "may potentially" said nothing a reader could use.

The rewrite is shorter, but length is not the point. The second version can be disagreed with, and the first cannot.

RULES

Writing rules

• 01 Find the action, make it the verb
• 02 Doer first, active voice
• 03 Short sentences, clear links
• 04 A banned list
• 05 Qualify only where it matters

01Find the action Lanham's paramedic method does most of this in one pass: find the action, make it the verb, put the doer first, cut the slow wind-up and the repeats (Lanham, 2006). Turning nouns back into verbs is Sword's (2012) fix for zombie nouns.

02Doer first Active voice tells the reader who acts (Department of the Army, 2020).

03Short sentences One idea per sentence, with the link to the next made explicit.

04A banned list Keep a list of words you reach for when you are not thinking: circle back, close the loop, whereby. Search the draft for them.

05Qualify where it matters A qualifier earns its place when it changes the meaning. In disagreement, also say whether the point is about evidence or values.

The paramedic method is the fastest single pass. Most drafts lose a quarter of their length to it and none of their content.

A DISTINCTION

Principles transfer. Preferences do not.

Teach the reasoning. Let people build their own house style.

Claim first, calibrated confidence and logical order hold for everyone. No serial commas, a punchy rhythm and a favourite palette are one writer's preferences.

Separate the two when you teach, or people copy your surface and miss your reasons.

A team lead marked every serial comma in a report as an error. The writers learnt to delete commas and nothing about putting the claim first. The next report had no serial commas and no claim.

08

SHAPE

Part four of five

08

Build for three depths of attention

The glance, the skim and the study each need their own surface.

The same piece of work is met in ten seconds by most readers, in ten minutes by some and in an hour by a few. One surface cannot serve all three.

This section covers depths of attention, when to write a memo instead of a deck and when a picture earns its place.

THREE DEPTHS

The glance, the skim, the study

Ten seconds, ten minutes, an hour. Each needs its own surface.

FIG 08 · ONE SOURCE, THREE SURFACES

One talk shipped as a 21-slide presenting deck and a 27-slide handout that carried the full argument and references. A primer was built so the slides stood alone without the document.

This file is the same idea: the deck, the longform and the posters come from one source, so they cannot drift apart.

IN ACTION · ILLUSTRATIVE

One evaluation, three surfaces

THE GLANCE

Ten seconds

• The subject line and one sentence: "Keep the reading pilot. Extend it to Year 3."

THE SKIM

Ten minutes

• Two pages: the claim, three findings with their confidence, the cost and the ask

THE STUDY

An hour

• The full report: method, data, limits and the annexes

An invented evaluation, written once and surfaced three times. Most readers stop at the glance, so the glance has to carry the recommendation. The adviser reads the skim. One analyst reads the study and checks the numbers.

Each surface is complete for its reader. None of them is a teaser for the next.

FORM

Slides for the room. Narrative for the decision.

DECK

Performed

• Sequenced
• Needs a speaker
• Good for the room

MEMO

Read

• Complete
• Survives without you
• Good for the decision

Some organisations ban slides for decisions and require a six-page narrative memo, read in silence at the start of the meeting, on the argument that full sentences force clearer thinking than bullets (Bezos, 2018). Tufte (2006) made the case against bullet hierarchies after the Columbia accident investigation.

Match the form to whether the reasoning must be read or performed.

A team took a two-page memo to a funding meeting instead of a deck. The panel read it in silence for the first ten minutes. The questions that followed were about the risks, not about what the slides meant.

PICTURES

A picture earns its place when it does work words cannot.

Decoration costs attention.

Diagrams help when they let the reader find related information by where it sits instead of by searching for it (Larkin & Simon, 1987). Interesting but irrelevant additions reduce learning, across a meta-analysis of the seductive-details literature (Rey, 2012).

Ask of every picture what the reader can see in it that the sentence beside it does not say.

RULES

Rules for slides and figures

• 01 Cut decoration
• 02 Signal the claim
• 03 Don't read the slide aloud
• 04 Label on the thing
• 05 One colour, one meaning

01Cut decoration Cut what does not serve the claim (Mayer, 2020).

02Signal the claim Make the important part visibly important.

03Do not read it aloud Speaking the words on screen competes with reading them (Mayer, 2020).

04Label on the thing Put labels on the marks they name, not in a legend the eye has to visit.

05One colour, one meaning One design primer gives one colour each to do, don't, explanation and evidence. Narrative also helps non-experts follow and remember, and it makes inaccurate claims more persuasive, so keep the evidence attached (Dahlstrom, 2014).

Most of these come from Mayer's (2020) multimedia principles. They are as true of a figure in a paper as of a slide.

THE ANNEX

Put the evidence in the annex so the body can carry the argument.

The body argues. The annex proves.

Full annexes belong in the reading version, with one pointer in the deck. Readers who need the proof know where it is, and readers who do not are not made to wade through it.

This document does the same. The appendices after the close carry the templates, the worksheets and the verification register.

A 60-page evaluation became a six-page paper and three annexes: the method, the data tables and the interview themes. The decision went through on the six pages. One adviser read the second annex and asked the only hard question, which the annex answered.

09

LAND

Part four of five

09

Brief before the room, check after it

A document informs the undecided. Relationships move the decided.

A finished paper is not yet a decision. Most decisions are shaped before the meeting that confirms them, and many are misunderstood after it.

This section covers what a document can and cannot do, readers who already disagree, timing, the room itself and checking what was heard.

WHAT A DOCUMENT CAN DO

A policy brief formed opinions in readers who had none. It barely moved readers who had one.

A randomised trial sent 807 readers different versions of one policy brief. The brief helped people with no prior view form one in line with the evidence but did little to shift strong priors. An attached opinion piece changed what people did more than what they believed (Beynon et al., 2012).

A systematic review of evidence use by policymakers found timely access and relationships with researchers among the most reported facilitators (Oliver et al., 2014).

WHEN THEY ALREADY DISAGREE

Show you understand them before you argue

Their position and where it holds. Yours and where it holds. What both sides gain.

01

Their position

• And where it holds

02

Your position

• And where it holds

03

What both gain

• The shared ground

Rogerian argument, adapted from Carl Rogers by Young, Becker and Pike (1970), aims to show readers they are understood, mark where their position is valid and establish shared values before proposing a solution (Purdue Online Writing Lab, n.d.-e). It is the written answer to the reader with a strong prior whom a brief alone will not move.

It fails against bad faith, and it can trade away a position that is simply right. Use it for disputes at the quality and policy stases, not for settled facts.

BEFORE THE ROOM

The meeting confirms what was settled in the corridor

Brief the adviser before the decider, and the decider before the room.

FIG 09 · WHERE THE DECISION IS ACTUALLY SHAPED

Nemawashi is preparing the ground, one conversation at a time, before the formal decision (Liker, 2004). Brief the adviser before the decider, and the decider before the room.

A pre-read that arrives the night before is a courtesy, not a briefing.

A proposal to merge two programmes was walked through with each director, one to one, in the fortnight before the meeting. Two objections came up and were fixed in the paper. The meeting took twelve minutes.

TIMING

The right paper at the wrong time is the wrong paper.

Policy moves when a problem, a solution and the politics line up (Kingdon, 2011). A paper that arrives before its window opens is filed, and one that arrives after it closes is history.

Think pieces can answer this from the other side. Horizon bands of one to three, three to five and five or more years, with a two-year revisit date, let a paper written early be picked up when its window opens.

A think piece on assessment and AI was written two years before anyone asked for it, with a revisit date on the cover. When a review was announced, the paper was already on the table and its framing shaped the review's terms of reference.

IN THE ROOM

In the room

• 01 Open with the ask
• 02 Answer their questions, not your slides
• 03 Clarify before you defend
• 04 Which stasis are we stuck at?
• 05 Steelman, then answer
• 06 Close on who does what

01Open with the ask Open with the decision you want.

02Their questions Present to their questions, not your slides. Skip what they already accept.

03Clarify first Separate clarifying questions from objections, and answer the first kind before treating anything as the second.

04Find the stasis When challenged, find where the disagreement sits: fact, definition, quality or policy (Purdue Online Writing Lab, n.d.-g). The shorter form is evidence or values.

05Steelman State the other side at its strongest before answering it. The Working Defaults field guide treats the principle of charity in full.

06Close on actions Close by saying what was decided and who does what.

The room rewards preparation done outside it. Most of these moves only work if the corridor conversations have already happened.

A WARNING

Silence is not agreement.

Ask each person directly.

Groups can agree to something nobody wants because each person assumes the others want it (Harvey, 1974). People speak up when they believe it is safe to be wrong in front of others (Edmondson, 1999).

The Collective Fallacies field guide treats the Abilene paradox in full.

A chair asked "Any objections?" and heard none. A week later two members said they had expected the proposal to fail without their help. Asking each member by name for one concern would have surfaced both.

IN ACTION · ILLUSTRATIVE

The playback that caught a misreading

SENT

What the paper said

• "Roll out the redesign now. Hold large schools until the class-size test is done."

HEARD

What the adviser played back

• "The redesign is not ready yet."

REVISED

The new first line

• "Roll out now in schools under 1,200 pupils. Large schools follow in Term 4."

The case is invented and the pattern is not. The paper led with a condition, and the adviser carried the condition to the decider instead of the recommendation. Nobody would have known without asking the adviser to say it back.

The fix was one sentence: put the action first and turn the exception into a date.

PLAYBACK

It is communicated when they can play it back

Ask the decider's adviser what they will tell the decider.

DO

Habits that travel

• Ask them to say it back
• Brief the adviser first
• Close by stating who does what

DO NOT

Habits that stall

• Assume the email was read
• Treat silence as consent
• Save the ask for the last slide

Teach-back in medicine asks patients to explain the plan in their own words, then fixes the gaps. Schillinger et al. (2003) linked this loop to better outcomes for patients with low health literacy. Craft guide 20 applies it to schools as explain-back.

For work communication, ask the decider's adviser what they will tell the decider.

Part five of five

Make it yours

Your own work, then your own standards.

• ONE Why work fails to travel
• TWO Decide what to say
• THREE Earn belief
• FOUR Make it land
• FIVE Make it yours

The last part turns the moves on your own work, then on your own standards.

10

PRACTISE

Part five of five

10

Do it to your own work

Bring one piece. Leave with a better one.

The moves in this session are easy to agree with and hard to apply to your own draft, because the author is the person least able to see what the draft leaves out.

This section turns the moves into a critique routine, a fifteen-minute exercise on a piece of your own work and a habit for keeping the standards you set.

CRITIQUE

Critique in passes, one job per pass

Before the first pass, run a premortem: it is six months later and this paper was rejected. Why?

PASS 1

Logic

• Does the claim follow?

PASS 2

Reader

• Can they act on it?

PASS 3

Evidence

• Does it trace down?

PASS 4

Words

• Can anything go?

Fix higher-order concerns before lower-order ones: the claim, the reader, organisation and development first, then sentences and spelling. One exception: errors in the first lines can stop a reader before they reach the claim (Purdue Online Writing Lab, n.d.-a). One pass trying all four finds the typos and misses the argument.

A premortem imagines the failure and surfaces reasons people already hold but will not volunteer (Klein, 2007). Critique against the goal, not against taste.

A reviewer read a draft four times: once for the claim, once as the director, once tracing every number to its source and once with a pen for cuts. The third pass found a figure from an older dataset that a single read had missed twice.

A CAUTION

Feedback can make work worse

0.41

Average effect of feedback on performance (d)

1 in 3

Feedback interventions that lowered performance

Across 607 effect sizes, feedback improved performance on average, but more than a third of feedback interventions made it worse, especially when the feedback drew attention to the person instead of the task (Kluger & DeNisi, 1996).

Critique the work, name the goal and propose the fix.

STANDARDS

If you keep fixing the same thing, write it down.

Your standards should outlive your attention.

A personal voice guide covers how you sound: sentence rhythm, the words you use and the words you never use, openings and closings. A design system means each visual decision is made once. Think-piece conventions let a whole group write to one standard. A decision log lets the next round start where this one stopped.

One reviewer kept fixing "utilise" and "in order to" in every draft. They wrote a one-page house list and shared it. Within a month the drafts arrived without them.

THE EXERCISE · 15 MINUTES

One piece of your own work

• 01 Write the reader, the act and the date
• 02 Write the claim in 25 words. Mark its stasis.
• 03 Reverse outline. Rewrite three headings as claims.
• 04 Toulmin your weakest claim. Underline the warrant.
• 05 Put a confidence bin on three claims
• 06 Run the paramedic method on your opening
• 07 Cut 30 per cent

01Two minutes Name the reader, the act as a verb and the date it is needed by.

02Two minutes Write the claim in 25 words or fewer. Mark whether it sits at fact, definition, quality or policy.

03Three minutes Write a five-to-ten-word summary beside each paragraph. Rewrite three headings as claims.

04Two minutes Lay out your weakest claim as grounds, warrant, claim, qualifier and rebuttal. Underline the warrant.

05Two minutes Mark three claims as established, emerging or extrapolated.

06Two minutes Find the action in each sentence of the opening paragraph, make it the verb and put the doer first.

07Two minutes Cut the draft by 30 per cent. Notice what you did not miss.

Bring a real piece: a paper, a brief, a deck or an evaluation you are working on now. The exercise works best on a draft that is nearly done, because that is where the author is most attached.

PAIRS

Pair critique: one job each

PARTNER A

Logic reader

• Reads only for the argument
• Reports one thing to cut
• Reports one thing to make explicit

PARTNER B

The named reader

• Reads as the person who must act
• Predicts where the paper goes next
• A wrong forecast marks a gap

Partner B reads as the named reader and, at each section break, predicts where the paper goes next. A wrong forecast marks a gap in the structure (Purdue Online Writing Lab, n.d.-a).

Each partner reports one thing to cut and one thing to make explicit. Two observations each are enough to act on.

THE FOLLOW-UP

Bring the revised piece back in two weeks.

One session teaches the idea. The second one changes the habit.

The evidence on professional learning argues for learning spread over weeks, worked on the participant's real material and followed up instead of certified (craft guide 21). A one-off workshop on communication would contradict the evidence it rests on.

THE PAIR, AGAIN

Which one did you bring in today?

A is a record of the work. B is the work, finished.

FIG 10 · THE OPENING PAIR, BACK ON SCREEN

The two openings from the start of the session. Look at the piece you brought and ask which of the two it resembles.

If it is A, the work is not wrong. It is unfinished, and the moves in this session are how to finish it.

If they have to find the point, you have not made one.

If you don't say how sure you are, they will assume you are surer than you are.

If you keep fixing the same thing, write it down.

THE CLOSE

The work is done when someone else can use it.

Claim first. Show how it was built. Say how sure you are. Check what they heard.

• AUTHOR geraldajam.com
• WRITING Purdue OWL

GOOD WORK DOES NOT SPEAK FOR ITSELF · 2026

Minto (2009) remains the clearest account of claim-first structure. Toulmin (1958) and the Purdue Online Writing Lab's pages on stasis and Rogerian argument carry the construction tools. For calibration, start with van der Bles et al. (2020).

The appendices in the longform version carry templates for the Vee, the Toulmin layout, a confidence vocabulary, a one-page decision submission and the stasis worksheet.

Reference layer

Appendix

Appendix A

The Vee, for a piece of work

Fill it before you write. Every claim on the doing side should trace down to a record. The worked example is a fictional pilot.

Appendix B

The Toulmin layout

One page per claim. Use it to build your own argument, and a second time to lay out someone else's before appraising it, such as a vendor's evidence of impact.

1Claim One sentence. What you want the reader to accept.

2Grounds Numbered facts, each with its source.

3Warrant In plain words, and underlined: why the grounds support the claim.

4Backing What makes the warrant credible.

5Qualifier How strong the claim is: a word from Appendix C or a number.

6Rebuttal An explicit "unless": the condition under which the claim fails.

Appendix C

A confidence vocabulary

The IPCC likelihood scale (Mastrandrea et al., 2010), with the three evidence bins. Proposed house rule: use a word from this list or a number, never a vague word on its own.

Appendix D

The one-page decision submission

A form for the short executive submission. One page, in this order.

1Decision sought and by when The act as a verb, and the date.

2The claim One sentence the decider could repeat.

3Why now What makes this the right window.

4Options Including doing nothing, with the cost of each.

5Recommendation Which option, and why.

6Confidence How sure you are, in the terms of Appendix C, and what would change your mind.

7Risks What could go wrong, and what you would do about it.

8Consulted Who has seen it, and where they stand.

9Annexes What is attached, for readers who need the proof.

Appendix E

Rewrite clinic

Three before-and-after pairs, from the smallest surface to the largest. All are invented for this appendix.

1An email subject Before: "Update on assessment module review" After: "Decision by 14 March: approve marking-flow redesign?" What changed: the act and the date moved into the subject. Move: aim.

2A slide heading Before: "Findings" After: "Teachers abandon the rubric step more than any other" What changed: a topic became a claim. Move: structure.

3A paper opening, through the paramedic method Before: "It is recommended that consideration be given by the committee to the adoption of a redesigned marking flow, as there has been an observed tendency on the part of teachers towards abandonment of the rubric step." Find the action: adopt, abandon. Put the doer first and make the action the verb: the committee adopts, teachers abandon. Cut the wind-up: "it is recommended that consideration be given". After: "We recommend the committee adopt a redesigned marking flow. Teachers abandon the rubric step more than any other." Move: write.

Appendix F

The eight tests

One question per move. Keep this beside the draft.

1Aim What can the reader do after reading this?

2Frame Could I state the claim before the evidence?

3Structure Do the headings alone tell the story?

4Construct Can a sceptic see how this claim was built?

5Calibrate Have I said how sure I am, in numbers?

6Write Would anything be lost if I cut this?

7Shape What does a reader get in ten seconds?

8Land Could the reader say it back to me?

Appendix H

The stasis worksheet

Before writing, and before the meeting. For each level, note what you claim, what the reader believes now and the evidence that would settle it. Mark the lowest level where you and the reader disagree. That is where the paper has to start (Purdue Online Writing Lab, n.d.-f).

Appendix I

The Rogerian layout

For a paper going to readers who already disagree (Purdue Online Writing Lab, n.d.-e). Not for bad faith, and not for settled facts.

1The issue Stated neutrally, in terms both sides accept.

2Their position Stated so they would recognise it.

3Where it holds The conditions under which they are right.

4Your position Stated as plainly.

5Where it holds The conditions under which you are right.

6What both gain The shared ground, and the proposal that builds on it.

Appendix G

Verification register

What was still open when this edition was built. Printed in the open, as the session recommends.

Source trail

References

The works this session draws on, and the best places to read further. Entries follow APA 7th edition. Links point to the publication or source page where one exists. Entries without a link are books or sources with no stable public address. The examples in the session are anonymised and may be composite or invented. They are illustrations, not sources, and are not listed here.

1. Aippersbach, S., Alley, M., & Garner, J. (2013, June). How slide design affects a student presenter's understanding of the content [Paper presentation]. ASEE Annual Conference & Exposition, Atlanta, GA, United States. https://doi.org/10.18260/1-2--19685
2. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Lawrence Erlbaum.
3. Beynon, P., Chapoy, C., Gaarder, M., & Masset, E. (2012). What difference does a policy brief make? Institute of Development Studies and International Initiative for Impact Evaluation.
4. Bezos, J. P. (2018). 2017 letter to shareholders. Amazon.com.
5. Bitzer, L. F. (1968). The rhetorical situation. Philosophy & Rhetoric, 1(1), 1–14.
6. Brizee, H. A. (2008). Stasis theory as a strategy for workplace teaming and decision making. Journal of Technical Writing and Communication, 38(4), 363–385. https://doi.org/10.2190/TW.38.4.d
7. Camerer, C., Loewenstein, G., & Weber, M. (1989). The curse of knowledge in economic settings: An experimental analysis. Journal of Political Economy, 97(5), 1232–1254. https://doi.org/10.1086/261651
8. Dahlstrom, M. F. (2014). Using narratives and storytelling to communicate science with nonexpert audiences. Proceedings of the National Academy of Sciences, 111(Suppl. 4), 13614–13620. https://doi.org/10.1073/pnas.1320645111
9. Department of the Army. (2019). Mission command: Command and control of Army forces (ADP 6-0).
10. Department of the Army. (2020). Preparing and managing correspondence (AR 25-50).
11. Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999
12. Flower, L. (1979). Writer-based prose: A cognitive basis for problems in writing. College English, 41(1), 19–37. https://doi.org/10.58680/ce197916016
13. Garner, J., & Alley, M. (2013). How the design of presentation slides affects audience comprehension: A case for the assertion–evidence approach. International Journal of Engineering Education, 29(6), 1564–1579.
14. Gopen, G. D., & Swan, J. A. (1990). The science of scientific writing. American Scientist, 78(6), 550–558.
15. Gowin, D. B. (1981). Educating. Cornell University Press.
16. Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics: Vol. 3. Speech acts (pp. 41–58). Academic Press.
17. Harvey, J. B. (1974). The Abilene paradox: The management of agreement. Organizational Dynamics, 3(1), 63–80. https://doi.org/10.1016/0090-2616(74)90005-9
18. Hinds, P. J. (1999). The curse of expertise: The effects of expertise and debiasing methods on predictions of novice performance. Journal of Experimental Psychology: Applied, 5(2), 205–221. https://doi.org/10.1037/1076-898X.5.2.205
19. Kent, S. (1964). Words of estimative probability. Studies in Intelligence, 8(4), 49–65.
20. Kingdon, J. W. (2011). Agendas, alternatives, and public policies (2nd ed.). Longman.
21. Klein, G. (2007). Performing a project premortem. Harvard Business Review, 85(9), 18–19.
22. Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119(2), 254–284. https://doi.org/10.1037/0033-2909.119.2.254
23. Lanham, R. A. (2006). Revising prose (5th ed.). Pearson Longman.
24. Larkin, J. H., & Simon, H. A. (1987). Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1), 65–100. https://doi.org/10.1111/j.1551-6708.1987.tb00863.x
25. Liker, J. K. (2004). The Toyota way: 14 management principles from the world's greatest manufacturer. McGraw-Hill.
26. Masset, E., Gaarder, M., Beynon, P., & Chapoy, C. (2013). What is the impact of a policy brief? Results of an experiment in research dissemination. Journal of Development Effectiveness, 5(1), 50–63. https://doi.org/10.1080/19439342.2012.759257
27. Mastrandrea, M. D., Field, C. B., Stocker, T. F., Edenhofer, O., Ebi, K. L., Frame, D. J., Held, H., Kriegler, E., Mach, K. J., Matschoss, P. R., Plattner, G.-K., Yohe, G. W., & Zwiers, F. W. (2010). Guidance note for lead authors of the IPCC Fifth Assessment Report on consistent treatment of uncertainties. Intergovernmental Panel on Climate Change.
28. Mauboussin, A., & Mauboussin, M. J. (2018, July 3). If you say something is "likely," how likely do people think it is? Harvard Business Review.
29. Mayer, R. E. (2020). Multimedia learning (3rd ed.). Cambridge University Press.
30. Minto, B. (2009). The pyramid principle: Logic in writing and thinking (3rd ed.). Pearson Education.
31. Mitchell, R. K., Agle, B. R., & Wood, D. J. (1997). Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of Management Review, 22(4), 853–886. https://doi.org/10.2307/259247
32. Novak, J. D., & Gowin, D. B. (1984). Learning how to learn. Cambridge University Press.
33. Oliver, K., Innvær, S., Lorenc, T., Woodman, J., & Thomas, J. (2014). A systematic review of barriers to and facilitators of the use of evidence by policymakers. BMC Health Services Research, 14, Article 2. https://doi.org/10.1186/1472-6963-14-2
34. Oppenheimer, D. M. (2006). Consequences of erudite vernacular utilized irrespective of necessity: Problems with using long words needlessly. Applied Cognitive Psychology, 20(2), 139–156. https://doi.org/10.1002/acp.1178
35. Patton, M. Q. (2008). Utilization-focused evaluation (4th ed.). Sage.
36. Purdue Online Writing Lab. (n.d.-a). Higher order concerns (HOCs) and lower order concerns (LOCs). Purdue University. https://owl.purdue.edu/owl/general_writing/mechanics/hocs_and_locs.html
37. Purdue Online Writing Lab. (n.d.-b). Organization and the CARS model. Purdue University. https://owl.purdue.edu/owl/general_writing/the_writing_process/organization_CARS_Model.html
38. Purdue Online Writing Lab. (n.d.-c). Paramedic method: A lesson in writing concisely. Purdue University. https://owl.purdue.edu/owl/general_writing/academic_writing/paramedic_method.html
39. Purdue Online Writing Lab. (n.d.-d). Reverse outlining: An exercise for taking notes and revising your work. Purdue University. https://owl.purdue.edu/owl/general_writing/the_writing_process/reverse_outlining.html
40. Purdue Online Writing Lab. (n.d.-e). Rogerian argument. Purdue University. https://owl.purdue.edu/owl/general_writing/academic_writing/historical_perspectives_on_argumentation/rogerian_argument.html
41. Purdue Online Writing Lab. (n.d.-f). Stasis theory. Purdue University. https://owl.purdue.edu/owl/general_writing/the_writing_process/stasis_theory/index.html
42. Purdue Online Writing Lab. (n.d.-g). Stasis theory for teamwork. Purdue University. https://owl.purdue.edu/owl/general_writing/the_writing_process/stasis_theory/stasis_for_teamwork.html
43. Rey, G. D. (2012). A review of research and a meta-analysis of the seductive detail effect. Educational Research Review, 7(3), 216–237. https://doi.org/10.1016/j.edurev.2012.05.003
44. Rozenblit, L., & Keil, F. (2002). The misunderstood limits of folk science: An illusion of explanatory depth. Cognitive Science, 26(5), 521–562. https://doi.org/10.1207/s15516709cog2605_1
45. Schillinger, D., Piette, J., Grumbach, K., Wang, F., Wilson, C., Daher, C., Leong-Grotz, K., Castro, C., & Bindman, A. B. (2003). Closing the loop: Physician communication with diabetic patients who have low health literacy. Archives of Internal Medicine, 163(1), 83–90. https://doi.org/10.1001/archinte.163.1.83
46. Swales, J. M. (1990). Genre analysis: English in academic and research settings. Cambridge University Press.
47. Sword, H. (2012). Stylish academic writing. Harvard University Press.
48. Toulmin, S. E. (1958). The uses of argument. Cambridge University Press.
49. Tufte, E. R. (2006). The cognitive style of PowerPoint: Pitching out corrupts within (2nd ed.). Graphics Press.
50. van der Bles, A. M., van der Linden, S., Freeman, A. L. J., & Spiegelhalter, D. J. (2020). The effects of communicating uncertainty on public trust in facts and numbers. Proceedings of the National Academy of Sciences, 117(14), 7672–7683. https://doi.org/10.1073/pnas.1913678117
51. Weiss, C. H. (1979). The many meanings of research utilization. Public Administration Review, 39(5), 426–431. https://doi.org/10.2307/3109916
52. Young, R. E., Becker, A. L., & Pike, K. L. (1970). Rhetoric: Discovery and change. Harcourt, Brace & World.

End of the session notes

Back to geraldajam.com