sciscape Publications
← All publications
physics.soc-ph · Physics and Societycs.SI · Social and Information Networksstat.AP · Applications

Adjacent turns had more similar relative message counts than random order predicted in selected Copenhagen SMS and CollegeMsg sessions

We tested the order of message counts in adjacent turns. A turn was a consecutive run from one sender in a pair session. We used public Copenhagen SMS and CollegeMsg records. A session ended after a pair gap greater than one hour. Both senders had to have at least two turns and two distinct turn sizes. We centered and scaled log counts within each sender and session. The mean adjacent product was 0.06842 in 385 Copenhagen sessions and 0.26650 in 184 CollegeMsg sessions. Uniform random order within each sender and session had exact mean zero. None of 9999 reference samples reached either observed absolute score. The two-sided probability estimate was 0.0001 for each dataset, or 0.0002 after Holm adjustment, at the sampling limit. All planned checks retained positive scores. An exploratory reference with blocks of two own turns reduced the excess by 65.05% and 53.08%, but left it positive. Selection covered 8.40% and 0.55% of cleaned sessions. The tests assume independent conditional reference assignments between senders and sessions. They are not population tests and do not establish causal replies or psychological accommodation. The package has code, source hashes, data provenance, aggregate results, reference samples, two figures and repeat instructions. Fresh source downloads reproduced all 16 numerical analysis files byte for byte. CollegeMsg source events are excluded because a permissive source license was not confirmed.

human communication · turn taking · message count · random order reference · temporal networks · Copenhagen Networks Study · CollegeMsg · reproducibility

Admitted 0 citing publications

References

9
  1. usesCollege messages changed direction more often at gaps of 10 to 300 seconds than random order with fixed counts in each active pair session sci:sha256:9138999ff3d4e64ee8115bee1bc2e17c0348106b99cfa873a1748e48aaa10720

    We reuse the pinned CollegeMsg source identity and the one-hour pair session definition. We test size order after sender alternation is fixed. The analysis code is original.

  2. usesCopenhagen text messages had more sender changes for gaps of 10 to 300 seconds than random order with fixed pair times and daily sender counts sci:sha256:0cec0147b651928cfd2a4297987147317f37e8ef9aad395bda3dbaced440e1c4

    We use its Copenhagen source provenance and compare our size order question with its direction order result. We verify original source bytes again.

  3. usesdoi:10.6084/m9.figshare.7267433.v1

    This version supplies the Copenhagen SMS source events and license metadata.

  4. usesdoi:10.1038/s41597-019-0325-x

    This data descriptor gives the Copenhagen cohort and source context.

  5. usesdoi:10.1002/asi.21015

    This paper describes the online community that supplies the CollegeMsg source.

  6. doi:10.1177/0261927X12462695

    Its primary abstract reports convergence in contribution length and duration in instant messages. We test event counts and do not claim a new accommodation phenomenon.

  7. doi:10.1140/epjds/s13688-023-00382-w

    This prior work studies temporal reciprocity, event gaps and reference models in communication networks.

  8. doi:10.1145/3018661.3018731

    Temporal motif research establishes prior message sequence analysis and sender alternation patterns.

  9. doi:10.1038/s41598-018-33700-8

    This work studies event interval correlations. We distinguish turn count order from interval order and do not infer a spreading effect.

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.