How to read an evidence label
Lesson 4 of 7 in this module · about 7 minutes · Romantic communication
The idea
Every claim in this course carries one of five labels, which say how strong the evidence is for the claim as worded, and so what you may conclude from it.
Well supported. A meta-analysis, which combines many earlier studies, or several independent studies agree, with no serious failure to replicate. You may write “research consistently finds” and “on average, people”. Example: heterosexual men on average rate people in cross-sex interactions as more flirtatious than heterosexual women do, a moderate difference with a lot of overlap, supported by three meta-analyses in one paper 1. It says nothing about who is more accurate and covers heterosexual cross-sex judgments only 1.
Supported with caveats. Several studies agree, with stated limits such as student samples or lab settings. You may write “several studies find” and “tends to”, and you should name the main limit. Example: several studies find that shared humor and laughter go with more liking and closeness in first conversations between strangers 2. The conversations were brief lab meetings, and the dating-interest result is one study of 102 pairs that cannot show cause 2.
Plausible. One study, or indirect evidence from a nearby question. You may write “one study found” and “there is some evidence that”. It is never a basis for advice on its own. Example: in two studies from one paper, people judged flirting more accurately when it was absent than when it was present, which suggests real flirting is easier to miss, and flirting was defined by the people’s own report 3.
Speculative. An idea or argument that has not been directly tested. You may write “X argues” and “X suggests, but this has not been tested”. It is never used to support advice. Example: Jean Smith’s HOT APE, from her 2015 TEDxLSHTM talk, presents six signs of flirting 4. A talk is not a peer-reviewed publication, and we have not seen the 250 interviews its online description mentions 4. The course uses it only as a reminder of where to look.
Marketing claim. Something a seller says about their product, without published evidence. You may write “X says” and “X advertises”. You never repeat it without naming who said it, and you never imply it works. Example: Marni Kinrys says her OSA (Observe, Share, Ask) method has been taught to over 10,000 men, a reach that is self-reported and unverified 5. The App Store listing for an app called Casanova Rizz AI says users can get “8x” and “9x” more matches, without citing research 6. A marketing label does not say the product fails or the claim is false. It says no evidence has been shown.
Why it matters
A meta-analysis, a talk and an advertisement can sound alike but deserve different weight.
What it looks like
Example
A friend repeats three claims at dinner: heterosexual men on average rate cross-sex interactions as more flirtatious than heterosexual women do 1; Jean Smith’s talk lists six signs of flirting 4; an app listing promises “9x” more matches 6. You ask of each, “Who says so?”
The same moment, handled the other way
Counterexample
Same dinner. The next day you pass on the matches claim as fact, without saying it came from an app listing. You also treat the flirtation finding as true of the next woman you meet.
A seller’s claim is the seller’s, and an average is not a statement about her.
The one thing to remember
The rule
Name a claim’s label and who says it, and never read any label as a fact about one person.
A label is not an importance ranking: a Plausible claim can matter more to a decision than a Well supported one. In this course, a finding about a population is never a statement about her. And no label makes a cue permission: it never turns a cue, silence or a past yes into a reason to touch, to take a bigger step, or to persist after a no.
Outside the course, ask who says it and how they know, how many independent sources agree, who was studied, and whether the claim says more than the evidence. If one source stands behind it, or the people studied are unlike the situation you face, give it less weight, most of all when the claim comes from someone selling a technique.
Try it now
Item 1 of this lesson’s set: A meta-analysis on sharing.
In the real world (optional)
If you like, when you next meet a piece of dating advice, ask who says it and how they know, then pick the label the source itself supports.
What this doesn’t tell you
It does not mean that only Well supported claims are worth knowing: the course uses Plausible and Speculative claims openly, with their labels.
Why do we believe this?
Each numbered mark in the text opens the evidence behind that sentence. Every claim in this course carries one of five labels.
- Well supported
- A meta-analysis or several independent studies agree, with no serious failure to replicate.
- Supported with caveats
- Several studies agree, with stated limits such as student samples or lab settings.
- Plausible
- One study, or indirect evidence from a nearby question.
- Speculative
- An idea or argument that hasn't been directly tested.
- Marketing claim
- Something a seller says about their product, without published evidence.
Evidence: Well supportedAcross studies, heterosexual men on average rate people in cross-sex interactions as more flirtatious, seductive and promiscuous than heterosexual women do, a moderate difference with a lot of overlap.
Why this label
A meta-analysis or several independent studies agree, with no serious failure to replicate. Three meta-analyses reported that men rated cross-sex interactions higher on all three dimensions, strongest for female targets with observers of face-to-face interactions and for male targets in face-to-face interactions. This is a difference in ratings, which does not by itself show who is accurate.
What it does not show
- Heterosexual cross-sex judgments only
- The meta-analysis shows that ratings differ, not that either group is correct
- Averages hide wide overlap; many men rate lower than many women
- Included observers of interactions and participants in face-to-face interactions; sample details not checked
- Only the abstract was read
Who it applies to
A pattern across groups of people, not a statement about any one person.
Source
- La France, B. H., Henningsen, D. D., Oates, A., Shaw, C. M. (2009). Social–Sexual Interactions? Meta-Analyses of Sex Differences in Perceptions of Flirtatiousness, Seductiveness, and Promiscuousness, Communication Monographs.
Evidence: Supported with caveatsIn first conversations between strangers, shared humor and laughter go with more liking and closeness, and in one dating-interest study both a man's humor and a woman's responsive laughter went with her interest.
Why this label
Several studies agree, with stated limits such as student samples or lab settings. Experiments with strangers show that humor and shared humorous experiences are associated with liking and closeness. In opposite-sex stranger pairs, mutual laughter and the woman's dating interest were associated, and responsiveness mattered as well as joke-telling.
What it does not show
- Mixed designs: some correlational, some manipulated humor tasks; the closeness experiment used same-sex pairs.
- Brief lab interactions (10-12 minutes in the dating-interest study); participant demographics not checked in full text.
- Dating-interest result is one study (102 dyads) and cannot say humor caused interest.
Who it applies to
A pattern across groups of people, not a statement about any one person.
Sources
- Treger, S., Sprecher, S., Erber, R. (2013). Laughing and liking: Exploring the interpersonal effects of humor use in initial social interactions, European Journal of Social Psychology 43(6), 532-543.
- Fraley, B., Aron, A. (2004). The effect of a shared humorous experience on closeness in initial encounters, Personal Relationships 11(1), 61-78.
- Hall, J. A. (2015). Sexual Selection and Humor in Courtship: A Case for Warmth and Extroversion, Evolutionary Psychology 13(3).
Evidence: PlausibleIn two studies, flirting was judged more accurately in interactions where it did not occur than where it did, which suggests real flirting is easier to miss than its absence.
Why this label
One study, or indirect evidence from a nearby question. Hall and colleagues found that people judged no-flirting interactions more accurately than flirting ones, and that the base rate of flirting in what observers saw shifted their accuracy. Flirting by women was judged more accurately than flirting by men.
What it does not show
- Two studies in one paper from one lab
- Study 1 used stranger pairs and Study 2 observers of 1-minute clips; sample details not checked
- Flirting defined by self-report
- Not a basis for assuming that any specific person is or is not flirting
Who it applies to
A pattern across groups of people, not a statement about any one person.
Source
- Hall, J. A., Xing, C., Brooks, S. (2015). Accurately Detecting Flirting: Error Management Theory, the Traditional Sexual Script, and Flirting Base Rate, Communication Research.
Evidence: SpeculativeJean Smith presents six signs of flirting, summarised as H.O.T. A.P.E.: Humor, Open body language, Touch, Attention, Proximity and Eye contact.
Why this label
An idea or argument that hasn't been directly tested. Smith's list is a practitioner's mnemonic from a conference talk. Research on the individual cues is in the topics flirting-signals, flirting-detection, humor-teasing, gaze, proximity and touch; none of that is cited in the talk as read.
What it does not show
- TEDx talk, not a peer-reviewed publication
- The YouTube description (re-checked 2026-10-04) confirms six steps named H.O.T. A.P.E.; the expansion into Humor, Open body language, Touch, Attention, Proximity and Eye contact was read from a third-party transcript and was not re-checked against the video
- The description says her research was 250 face-to-face interviews in four cities, published as her book The Flirt Interpreter; that book was not read
Who it applies to
Depends on the setting and the people involved.
Source
- Smith, J. (2015). The Science of Flirting: Being a H.O.T. A.P.E. | Jean Smith | TEDxLSHTM, TEDx Talks (YouTube).
Evidence: Marketing claimMarni Kinrys describes OSA as a conversation method of observation, sharing and asking a question, which she says she created and has taught to over 10,000 men.
Why this label
Something a seller says about their product, without published evidence. Creator attribution and reach are self-reported. The structure overlaps self-disclosure-questions and responsiveness, which cover reciprocal disclosure and follow-up questions.
What it does not show
- Self-reported reach, unverified
- Step-by-step detail was not on the free pages read
Who it applies to
Depends on the setting and the people involved.
Sources
- Kinrys, M. (2013). How To Talk To Women Using OSA Formula | Wing Girl Method, winggirlmethod.com.
- Kinrys, M. (2016). Effortless Approach O.S.A : A Tool That Helps You Start Conversations With Women - Marni's Wing Girl Method, winggirlmethod.com.
- Kinrys, M. (2014). Great Opener To Say To Women - Marni's Wing Girl Method, winggirlmethod.com.
Evidence: Marketing claimThe Casanova Rizz AI app listing promises AI-written replies, profile analysis and photo tools, and claims users can get '8x' and '9x more matches', without citing research.
Why this label
Something a seller says about their product, without published evidence. App-store marketing. No evidence is offered for the match-rate claims. Identity with the spec's 'Casanova AI' is a name match only.
What it does not show
- Seller's claim
- Different products share the 'Casanova' name
Who it applies to
A pattern across groups of people, not a statement about any one person.
Source
- Casanova AI (undated). Casanova Rizz AI App - App Store, apps.apple.com.
More practice
The rest of the set, Evidence check: five labels: 4 items, about 4 minutes. Decide what the evidence supports, and how well. Open the exercise set.
Go further
Outside talks, videos and pages on the topics this lesson cites. Nothing from them loads until you follow a link, and a link is not an endorsement: read the note under each one.
Vanessa Van Edwards: Why Women Never Make The First Move!
Van Edwards on who tends to initiate in courtship and the small opening signals that precede an approach; read the video before linking to confirm the claims shown.
Keep in mind. Popular-science framing; check any quoted statistic against the primary study before the app repeats it.
Marni Kinrys: "You Think You're Pretty Smooth, Don't You?" She's Flirting. Not Being Mean.
Kinrys's short presents playful teasing and mock challenge as a sign of attraction and advises playing back rather than withdrawing. Relevant to engaged versus merely polite behaviour.
Keep in mind. Anecdotal; teasing can be friendly, polite or unkind, and is not by itself a sign of romantic interest. Check for reciprocity over several turns.
Jeffrey Hall (University of Kansas): Flirting hard to detect, researcher says
University summary of Hall's study of first conversations: partners were far less accurate at detecting flirting than at noticing its absence, and observers were no better.
Keep in mind. News release; student samples and one lab setting.