Table A2

Metrics

CodesMacro categoriesMetrics
1Amount of words, messages and characters1 = Word counts
2 = Message counts
3 = Type-token ratio (total number of different words)
4 = Shorthand
5 = Emoticons
6 = LIWC metrics (counts words)
2Tone of conversation/interaction with the chatbot7 = Profanity (reinaava)
8 = User attitude towards chatbot
9 = Message supportiveness
10 = Message effectiveness
11 = Sincerity
12 = Thoughtful (pleasantness)
13 = Polite (pleasantness)
14 = Pre-responsive (pleasantness)
15 = Friendly self-introduction
16 = Colloquial style
17 = Friendly farewell
3Perceived feelings (conversation with the chatbot)18 = Eeriness (fear)
19 = Belief in robotic feelings
20 = Felt supported
21 = Likeability
22 = Intelligence
23 = Attitude
24 = Novelty
25 = Spine-tingling perception
26 = Perceived interactivity
27 = Perceived contingency
28 = Perceived dialogue
29 = User engagement (level of absorption)
30 = Topic involvement
31 = Perceived sadness
32 = Perceived recognition
33 = Perceived understanding
34 = Belief in robotic intelligence
35 = User satisfaction
4Grammatical structures36 = Grammatical structures
37 = Vocabulary range
38 = Spelling
39 = Upper/lower case
40 = Nouns
41 = Pronouns
42 = Verbs
43 = Articles
44 = Question words
45 = Word order in statements
46 = Word order in questions
47 = Utterances
48 = Typos
49 = Capitalized words
5Chatbots interaction skills50 = Percentage of follow-up questions
51 = Number of coherent conversation turns
52 = Percentage of successful resolutions
53 = Gricean maxims (how speakers act cooperatively)
54 = Agreement
55 = Precision
56 = Recall
57 = Accuracy
58 = Task completion
59 = Use of context
60 = Correction rate
61 = Response satisfaction
62 = Conversation ability
63 = Skilled (conversational skill)
64 = Human (conversational skill)
65 = Engaging (conversational skill)
66 = Used words
67 = Used tones
68 = Themes
69 = Detecting emotions in textual dialogues
70 = Word -level features
71 = Phrase-level features
72 = Sentence level features
73 = Semantic-level features
74 = Turing test
6Conversation style (speed, habits, special features)75 = Quick answers
76 = Slow answers
77 = Mean answers
78 = Talks about self
79 = Questions about the agent
7Chatbot usage80 = Previous usage
81 = Power usage (attitude and use of technologies)

or Create an Account

Close Modal
Close Modal