- `b_*`: lasso coefs
- `b_debias_*`: debiased lasso coefs
- `*crowd*`: strategy presence is mean(deepseek)
```r
features b_crowd b_deepseek b_gpt b_debias_crowd b_debias_deepseek b_debias_gpt p_crowd p_deepseek p_gpt sig_crowd sig_deepseek sig_gpt
<char> <num> <num> <num> <num> <num> <num> <num> <num> <num> <char> <char> <char>
1: preZ -2.38 -1.53 -2.42 -3.07 -2.29 -3.08 0.00 0.00 0.00 *** *** ***
2: address_objections_and_counterarguments -3.15 -1.31 -2.26 -3.59 -2.17 -2.55 0.00 0.00 0.00 *** ** ***
3: audience_adaptation 2.11 0.00 1.68 3.38 1.50 2.21 0.00 0.12 0.01 ** **
4: topicZ 0.39 0.36 0.43 0.51 0.40 0.56 0.05 0.08 0.04 * . *
5: politeness_and_civil_tone 0.00 0.00 0.00 3.85 1.83 2.58 0.05 0.10 0.27 . .
6: contrastive_tone 1.42 0.07 1.99 1.86 0.72 2.47 0.06 0.26 0.00 . **
7: positive_tone 1.06 1.15 0.42 1.94 1.72 1.20 0.07 0.03 0.18 . *
8: conditionZ:preZ 0.33 0.26 0.30 0.38 0.32 0.42 0.09 0.13 0.06 . .
9: evidence_or_factbased_arguments 0.85 0.13 0.77 1.27 0.84 1.12 0.10 0.21 0.10
10: emotional_appeal_with_balanced_urgency 0.00 0.00 0.00 -1.70 -1.69 -0.36 0.11 0.06 0.67 .
11: conditionZ:topicZ -0.18 -0.09 -0.21 -0.36 -0.25 -0.36 0.12 0.26 0.13
12: encourage_action_with_clear_calls 0.15 0.00 0.57 1.22 0.62 1.23 0.16 0.30 0.10
13: transfer_of_association 0.40 0.00 0.66 1.85 0.43 1.08 0.18 0.70 0.20
14: conditionZ 0.00 -0.08 0.12 0.50 0.12 0.58 0.19 0.74 0.08 .
15: negative_tone 0.54 0.00 1.47 1.73 0.40 1.91 0.20 0.67 0.03 *
16: stimulate_anger 0.00 0.00 0.00 -4.55 -19.28 -1.51 0.21 0.16 0.23
17: relatable_hypotheticals 0.00 0.00 0.00 1.77 1.98 0.73 0.26 0.25 0.46
18: stimulate_enthusiasm 0.00 0.00 0.00 -0.95 -0.73 0.17 0.31 0.35 0.85
19: accuracyS 0.00 0.00 0.00 1.60 1.03 1.00 0.31 0.51 0.51
20: topicZ:preZ -0.13 -0.02 -0.19 -0.20 -0.17 -0.27 0.35 0.37 0.21
21: use_of_negative_testimonials 0.00 0.00 0.00 -1.85 -1.45 -1.07 0.54 0.65 0.58
22: namecalling 0.00 0.00 0.00 1.25 -10.34 1.10 0.58 0.39 0.28
23: cognitive_elaboration 0.00 0.00 0.00 -0.49 -0.79 -0.12 0.66 0.27 0.88
24: reciprocity_and_mutual_benefit 0.00 0.00 0.00 1.09 7.22 -1.25 0.67 0.16 0.28
25: positive_framing_and_value_alignment 0.00 0.00 0.00 -0.44 0.44 -0.53 0.68 0.56 0.48
26: build_rapport_and_common_ground 0.00 0.00 0.00 0.63 0.34 0.36 0.70 0.81 0.72
27: aggressive_and_explicit_directives 0.00 0.00 0.00 1.88 -9.14 1.70 0.70 0.05 0.51 *
28: active_listening_and_empathy 0.00 0.00 0.00 -0.37 -0.50 -0.21 0.73 0.51 0.80
29: use_of_positive_testimonials 0.00 0.00 0.00 -0.68 0.05 -0.68 0.75 0.98 0.73
30: localized_focus 0.00 0.00 0.00 -0.36 0.32 -0.39 0.80 0.83 0.72
31: social_proof_and_normative_influence 0.00 0.00 0.00 0.53 -1.34 1.10 0.83 0.67 0.40
32: storytelling_and_relatable_anecdotes 0.00 0.00 0.00 -0.23 -0.50 0.23 0.85 0.66 0.79
33: conditionZ:topicZ:preZ 0.00 0.00 0.00 -0.02 -0.03 0.00 0.94 0.89 0.98
34: use_of_everyday_people_as_messengers 0.00 0.00 0.00 0.39 -0.38 0.50 0.95 0.91 0.92
features b_crowd b_deepseek b_gpt b_debias_crowd b_debias_deepseek b_debias_gpt p_crowd p_deepseek p_gpt sig_crowd sig_deepseek sig_gpt
```
![[20250212150246.png]]
![[20250212150214.png]]