David P.B. Norris
The measure only exists where this member actually spoke in the debate their vote belongs to, and where that vote can be tied to a specific question. Both cuts are steep.
| Votes cast | 9 | |
| … spoke in that debate section | 7 | |
| … vote tied to a specific question | 5 | |
| … position stated clearly enough to score | 1 |
The third row is the sharpest cut. A committee-stage sitting can hold fifteen divisions while its speeches discuss forty-two numbered amendments, and the record does not say which amendment each division was. Those votes are counted here and deliberately left unscored rather than guessed.
The same scorable votes, split by whether the member went with their party. The top-right cell is the one worth reading: a stated position reversed while following the party line.
| Voted with own party | Voted against own party | |
|---|---|---|
| Matched their words | 1 | 0 |
| Contradicted their words | 0 | 0 |
Not scored: 2 where the vote could not be tied to a question, 0 where no clear position was stated in the debate.
Valence runs −1 (negative) to +1 (positive). This is a separate measure from alignment above and says nothing about how anyone voted — "the Minister has been asleep at the wheel" is strongly negative and carries no position on the question.
| Model | Speeches scored | Mean valence | Corpus mean |
|---|---|---|---|
| vader | 74 | +0.539 | +0.423 |
Mean tone by topic, most negative first. Topics come from an LDA model fitted over debate contributions; the terms are the model's, the labels are hand-assigned.
| Topic | Speeches | Mean valence |
|---|---|---|
| Health scandals, inquiries and bereaved families hospital · families · family · cancer · children · did | 8 | +0.402 |
| Procedural and rhetorical filler (low-content) important · like · good · debate · did · does | 32 | +0.627 |
| Party-political exchange (party names dominate) workers · rights · right · sinn · féin · irish | 6 | +0.930 |
Contributions whose tone sits more than two standard deviations from this member's own average. Baselines are per member on purpose — an absolute threshold would just rank people by temperament.
I appreciate that. I have come to Seanad Éireann to raise the question of cluster bombs. This is particularly appropriate since the Convention on Cluster Munitions, effectively banning them, was signed in Dublin in May 2008. Figures in 2018 show that more than…
I thank the Cathaoirleach. I was not in the House last week, so I did not get the opportunity to comment on the Pat Finucane affair. From my background as a southern unionist, I take it extremely badly that the British Government has shown itself to be totally…
The tobacco lobby has been absolutely disgraceful. It knew about the link between lung cancer and smoking and suppressed it. It had that knowledge, which should have been publicly available. That industry is an absolute and utter disgrace. It was in it for the…
I thank the Cathaoirleach. I have two matters to raise. The first is a question that has been drawn to my attention by a couple. The man is infertile as a result of Klinefelter syndrome and they got a friend of theirs to donate sperm. The woman got pregnant an…
Alignment by the party a member sat for on the day, with 95% Wilson intervals. This is the test the measure could have failed.
| Party | Aligned | 95% CI | n |
|---|---|---|---|
| PBP | 82.5% | 68–91% | 40 |
| LAB | 78.2% | 69–85% | 101 |
| IND | 76.5% | 72–81% | 327 |
| SF | 75.8% | 69–81% | 198 |
| SD | 71.4% | 58–82% | 49 |
| FG | 69.1% | 62–76% | 162 |
| FF | 67.9% | 60–75% | 165 |
| GP | 56.8% | 41–71% | 37 |
Members of government parties align their stated position with their vote 9.4 points less often than Independents — bootstrap 95% CI [+2.7, +16.0], which excludes zero. Opposition parties sit 9.5 points above government too, CI [+3.1, +15.9]. Independents and opposition members are indistinguishable from each other (−0.0 points, CI [−6.3, +6.2]). So the gap does not come from having no party; it comes from being in government, which is what a whip effect should look like. Read this as the measure behaving sensibly, not as a finding about any individual.
Two automated methods were built and both failed. What survives is the rule-based layer, which is why the coverage above is narrow.
| Method | Accuracy | 95% CI | n |
|---|---|---|---|
| nli-vs-mover-anchor:deberta | 35.9% | 29–42% | 234 |
| nli-vs-mover-anchor:distilroberta | 11.5% | 8–16% | 234 |
Zero-shot entailment was asked to judge contributions whose true position is known — the member moved the amendment, so the answer is support by construction — and got it right 11.5% of the time, returning "unclear" or "oppose" for nearly everything. A classifier trained on the rule-labelled cases over the corpus embeddings reached 71.1% against a 65.3% majority-class baseline: better than chance, not better enough. Both are excluded from every figure on this page. Only the rule-based cascade, which is exact where it fires and silent where it does not, is used.
Position is read from the member's own contributions in the debate section the division belongs to, by a three-step cascade: an explicit move of the question ("I move amendment No. 5", true by construction); declarative stance phrases ordered so negations are tested first, because "I cannot support" and "I do not oppose" both defeat a naive keyword match; and, where neither fires, a zero-shot entailment model judging the contribution against the question put. Tone uses a separate model and is reported separately. Agreement against hand-labelled cases, with bootstrap intervals, is on the evidence page; no score here should be read without it.