The file KUPA-KEYS-META.csv of demographic data, metadata, and pre-processed data extracted from the keylogger includes the following 38 columns:
id: The unique user id that participated in the survey. This is not connected with the participants' Prolific identification number. It is the submission number extracted from Qualtrics.
navigator_useragent: This value is extracted from the keylogger application and contains information about the user's web browser and device.
navigator_language: This value extracted from the keylogger application and contains information about the user's preferred language setting in their web browser. This setting is typically determined based on the user's browser preferences or system settings.
age: Response to the question "What is your age in years?". The value is an integer in [18,99].
handedness: This is the participants' response to the question "Are you right-handed, left-handed, or ambidextrous". The value is in the set {Right-handed, Left-handed, Ambidextrous}.
comphours: Response to the question "How many hours per day do you spend on a computer?". The value is an integer in [0,24].
layoutcomf: Response to the question "Which keyboard layout are you most comfortable with?". The value can be QWERTY, QWERTZ, AZERTY, IDK (I don't know), or Other (specified in a text box).
layoutnow: Response to the question "Which keyboard layout are you using now?". As previously, the value can be QWERTY, QWERTZ, AZERTY, IDK (I don't know), or Other (specified in a text box).
comptype: Response to the question "Are you using a desktop or a laptop now?". The value is in {Desktop, Laptop}. Note that in this survey we only allowed using a desktop computer or a laptop.
countryres: Response to the question "What is your current country of residence?". A full comprehensive list of countries was provided as a drop down list.
nativelang: Response to the question "What is your native language?". A full comprehensive list of countries was provided as a drop down list.
englishyears: Response to the question "How many years have you been learning English?". This question was provided to non-native English speakers only. The value is a float number.
englishcountrymonths: Response to the question "How much time (in months) have you spent in an English-speaking country over the last 3 years?". This question was provided to non-native English speakers only. The value is an integer in [0,36].
englishexposure: Response to the question "How many hours per day are you exposed to English on average?". This question was provided to non-native English speakers only. The value is an integer in [0,24].
otherlanguages: Response to the question "Besides your native language and English, what other languages do you speak? Choose the language you are most comfortable with". A full comprehensive list of countries was provided as a drop down list.
cefrself: Response to the question "Do you consider yourself a beginner (A1, A2), intermediate (B1, B2), or advanced (C1, C2) user of English?". This question was provided to non-native English speakers only. The value is in the set {'Advanced (C1, C2)', 'Beginner (A1, A2)', 'Intermediate (B1, B2)'}, or null in the case of native English speakers.
cefrlevel: Response to the question "If you have passed an English language exam, please let us know your highest level on the CEFR scale". The value is in the set {'I have not taken an English language exam', 'My CEFR level is A1 or A2', 'My CEFR level is B1 or B2', 'My CEFR level is C1 or C2'}.
cefrwhen: Response to the question "In which year did that assessment take place?". This question was provided only to participants who reported they've passed an English language exam in the previous question.
educ: Response to the question "What is the highest degree or level of education you have completed?". The options provided were "High School", "Bachelor's Degree (e.g., BSc, BA, MB, etc.)" "Master's Degree (e.g., MSc, MEng, MRes, etc.)", "Doctoral Degree or higher (e.g., PhD, MPhil, etc.)", or "Other (please specify)".
time: The total time (in seconds) the participants needed to complete the survey, including the two tasks.
task1_time: The time (in seconds) the participants spent in the copy-text task.
task2_time: The time (in seconds) the participants spent in the essay-writing task.
task1_events: The number of events recorded by the keylogger for the copy-text task. Note that this figure includes both user-related events and system-related events.
task2_events: The number of events recorded by the keylogger for the essay-writing task. Note that this figure includes both user-related events and system-related events.
task1_words: The number of words in the final submitted text for the copy-text task.
task2_words: The number of words in the final submitted text for the essay-writing task.
cohort: The cohort number the participant. There were three cohorts. For the first cohort, the minimum number of words for the essay-writing task was specified as 250. This cohort also includes native English speakers. For cohorts 2 and 3, the minimum number of words was reduced to 150. There is no other difference between these cohorts, other than the season of when the survey took place.
prompt_id: An integer specifying the prompt id for the essay-writing task. Each participant was randomly allocated to a prompt out of ten options.
prompt: The actual essay prompt (as text).
answer: The final submitted answer of the user in the text box.
mark_a0: The mark of the automarker (W&I) as an integer in [0,12] on the submitted essay.
mark_h1: The mark of the first human marker as an integer in [0,12] on the submitted essay.
mark_h2: The mark of the second human marker as an integer in [0,12] on the submitted essay.
mark_h3: The mark of the third human marker as an integer in [0,12] on the submitted essay.
cefr_a0: The CEFR level predicted by the automarker on the submitted essay.
cefr_h1: The CEFR level from the first human marker on the submitted essay.
cefr_h2: The CEFR level from the second human marker on the submitted essay.
cefr_h3: The CEFR level from the third human marker on the submitted essay.
For the copy-text task and the essay-writing task we provide two more tables, KUPA-KEYS-TASK-1.csv and KUPA-KEYS-TASK-2.csv. For each table, each row corresponds to an event and includes the 13 following columns:
id: The participant’s identification anonymised number. This is to be used for the correspondence between the keystroke log data and the entries in the table
KUPA-KEYS-META.csv.time: timestamp of the event, in milliseconds since the application was started.
type: the event type. The possible values are 'down' when a keyborard key is pressed, 'up' when a keyboard key is released, 'click' for a mouse click, 'insert' when the content of the text-box is updated, and 'capture', which is periodically triggered to save the current textbox state.
key: for key press and release events, this is the actual value of the key pressed.
key_code: for key press and release events, this is the name of the physical key on the keyboard rather than the specific layout chosen by the user. The values are associated with a US-based QWERTY layout.
alt_key: indicator of whether an alt key is also pressed at the time of the event.
ctrl_key: indicator of whether a control key is also pressed at the time of the event.
meta_key: indicator of whether a meta key is also pressed at the time of the event.
shift_key: indicator of whether a shift key is also pressed at the time of the event.
is_repeat: indicator of whether this is a key that is automatically repeating because the key is held down.
range_start: At the time of the event, the location as an integer offset in the text-box of either the cursor or the start of any selected text.
range_end: At the time of the event, the location as an integer offset in the text box of either the cursor or the end of any selected text.
text: the text in the text box when there is a capture event type. This may also be populated when there is an insert event type, with the content added to the text box.