This page documents every column in all_data, the pooled
dataset described on the sample
description page. It is meant to serve the same purpose as the data
dictionaries shared alongside the flat merged_data.xlsx
file archived on OSF, for anyone who wants
to reuse the data without also pulling in this package.
The tables below are computed directly from all_data
rather than typed out by hand, so they cannot drift out of sync with the
data itself. A machine-readable version of this same information,
following the Psych-DS
dataset_description.json convention, is archived alongside
the data on OSF.
Core variables
| Variable | Description | Type | Range / levels | N missing |
|---|---|---|---|---|
| study | Study of origin. | string | burns, monzel, mas, ruby, kvamme | 0 |
| lang | Language in which the study was administered. | string | en, fr | 0 |
| id | Unique participant identifier
(“subj_ |
string | subj_[study]_[number] | 0 |
| sex | Sex assigned at birth, as reported by the participant. | string | female, male, other | 1 |
| gender | Gender identity, as reported by the participant, where collected separately from sex. | string | female, male, other | 1 |
| age | Participant age in years, self-reported. | integer | 10-86 | 3 |
| vviq | Total score on the VVIQ (Marks, 1973), summed across all 16 items. Lower scores indicate weaker/absent visual imagery. | integer | 16-80 | 0 |
| tas | Total score on the TAS-20 (Bagby et al., 1994), summed across all 20 items (already reverse-scored where applicable). Higher scores indicate greater alexithymia. | integer | 20-94 | 0 |
| tas_identify | Difficulty Identifying Feelings (DIF) subscale of the TAS-20. Sum of items 1, 3, 6, 7, 9, 13, 14. | integer | 7-35 | 0 |
| tas_describe | Difficulty Describing Feelings (DDF) subscale of the TAS-20. Sum of items 2, 4, 11, 12, 17. | integer | 5-25 | 0 |
| tas_external | Externally Oriented Thinking (EOT) subscale of the TAS-20. Sum of items 5, 8, 10, 15, 16, 18, 19, 20. | integer | 8-36 | 0 |
| tas_group | Alexithymia status from the standard TAS-20 cutoff (total >= 61 = alexithymia). | string | alexithymia, typical_tas | 0 |
| vviq_group_4 | VVIQ group, 4 levels: aphantasia, hypophantasia, typical, hyperphantasia. | string | aphantasia, hypophantasia, typical, hyperphantasia | 0 |
| vviq_group_3 | VVIQ group, 3 levels: aphantasia, hypophantasia, typical (typical and hyperphantasia collapsed). | string | aphantasia, hypophantasia, typical | 0 |
| vviq_group_2 | VVIQ group, 2 levels: aphantasia, typical (broadest split). | string | aphantasia, typical | 0 |
VVIQ item-level responses
(vviq_q1-vviq_q16)
The VVIQ (Marks, 1973) has no subscales; its 16 items are grouped into 4 scenes of 4 items each (a relative or friend you often see; the rising sun; the front of a familiar shop; a country scene with trees, mountains, and a lake), each rated 1 (“perfectly clear and as vivid as normal vision”) to 5 (“no image at all”). Item wording is not reproduced here, as it is copyrighted by the scale’s author — see Marks (1973) for the original items.
| Variable | Scene | Type | Range / levels | N missing |
|---|---|---|---|---|
| vviq_q1 | a relative or friend you often see | integer | 1-5 | 105 |
| vviq_q2 | a relative or friend you often see | integer | 1-5 | 105 |
| vviq_q3 | a relative or friend you often see | integer | 1-5 | 105 |
| vviq_q4 | a relative or friend you often see | integer | 1-5 | 105 |
| vviq_q5 | the rising sun | integer | 1-5 | 105 |
| vviq_q6 | the rising sun | integer | 1-5 | 105 |
| vviq_q7 | the rising sun | integer | 1-5 | 105 |
| vviq_q8 | the rising sun | integer | 1-5 | 105 |
| vviq_q9 | the front of a familiar shop | integer | 1-5 | 105 |
| vviq_q10 | the front of a familiar shop | integer | 1-5 | 105 |
| vviq_q11 | the front of a familiar shop | integer | 1-5 | 105 |
| vviq_q12 | the front of a familiar shop | integer | 1-5 | 105 |
| vviq_q13 | a country scene with trees, mountains, and a lake | integer | 1-5 | 105 |
| vviq_q14 | a country scene with trees, mountains, and a lake | integer | 1-5 | 105 |
| vviq_q15 | a country scene with trees, mountains, and a lake | integer | 1-5 | 105 |
| vviq_q16 | a country scene with trees, mountains, and a lake | integer | 1-5 | 105 |
Item-level VVIQ data are available for the burns, mas, ruby, and
kvamme studies. They are missing (NA) for all 105 Monzel
participants, for whom only total scores were shared.
TAS-20 item-level responses
(tas_q1-tas_q20)
The TAS-20 (Bagby et al., 1994) has 20 items rated 1 (“strongly disagree”) to 5 (“strongly agree”), assigned to the three sub-scales described above. Item wording is not reproduced here, as it is copyrighted by the scale’s authors — see Bagby et al. (1994) for the original items.
Five items (4, 5, 10, 18, 19) are reverse-keyed on the original
scale. Values in tas_q4, tas_q5,
tas_q10, tas_q18, and tas_q19 in
all_data/merged_data.xlsx are already
reverse-scored — summing
tas_q1-tas_q20 directly reproduces the
tas total column, with no further reversal needed.
| Variable | Subscale | Reverse-keyed | Type | Range / levels | N missing |
|---|---|---|---|---|---|
| tas_q1 | DIF | integer | 1-5 | 105 | |
| tas_q2 | DDF | integer | 1-5 | 105 | |
| tas_q3 | DIF | integer | 1-5 | 105 | |
| tas_q4 | DDF | yes (already reversed) | integer | 1-5 | 105 |
| tas_q5 | EOT | yes (already reversed) | integer | 1-5 | 105 |
| tas_q6 | DIF | integer | 1-5 | 105 | |
| tas_q7 | DIF | integer | 1-5 | 105 | |
| tas_q8 | EOT | integer | 1-5 | 105 | |
| tas_q9 | DIF | integer | 1-5 | 105 | |
| tas_q10 | EOT | yes (already reversed) | integer | 1-5 | 105 |
| tas_q11 | DDF | integer | 1-5 | 105 | |
| tas_q12 | DDF | integer | 1-5 | 105 | |
| tas_q13 | DIF | integer | 1-5 | 105 | |
| tas_q14 | DIF | integer | 1-5 | 105 | |
| tas_q15 | EOT | integer | 1-5 | 105 | |
| tas_q16 | EOT | integer | 1-5 | 105 | |
| tas_q17 | DDF | integer | 1-5 | 105 | |
| tas_q18 | EOT | yes (already reversed) | integer | 1-5 | 105 |
| tas_q19 | EOT | yes (already reversed) | integer | 1-5 | 105 |
| tas_q20 | EOT | integer | 1-5 | 105 |
Item-level TAS-20 data are available for the same four studies as the VVIQ items above, and missing for the same 105 Monzel participants.
A note on sex and gender
For most participants, sex and gender
coincide. The Burns study collected both constructs separately, and a
small number of participants (n = 4) reported a sex and gender that
differ.
Continuing through the Extended Online Report: this page is a technical reference, meant to be partially independent from the rest of the vignettes. See the sample description page for how the five studies were combined, or get back to the home page.