https://github.com/ddbourgin/mturk_utils

Kludgey helper utils for running studies on Mechanical Turk
https://github.com/ddbourgin/mturk_utils

Last synced: about 1 month ago
JSON representation

Kludgey helper utils for running studies on Mechanical Turk

• Host: GitHub
• URL: https://github.com/ddbourgin/mturk_utils
• Owner: ddbourgin
• Created: 2018-05-20T00:37:37.000Z (about 8 years ago)
• Default Branch: master
• Last Pushed: 2018-06-01T07:45:12.000Z (about 8 years ago)
• Last Synced: 2025-01-25T10:42:19.385Z (over 1 year ago)
• Language: Python
• Size: 21.5 KB
• Stars: 4
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# mturk utils

Convenience scripts for interacting with MTurk/Psiturk. 

Before using the scripts below, define the following environment variables:

    > export AWS_ACCESS_KEY_ID=

    > export AWS_SECRET_ACCESS_KEY=

## approve_batch.py

**Usage:** `approve_batch.py [-h] [-t TITLE]`

Batch HIT approver. Only approves HITs that are listed as reviewable,

saving a log of subject and HIT IDs that it approves to the current directory in an `.npz` archive. Can be run multiple times as more HITs are posted.

#### Optional arguments

  - `-h`, `--help`            show help message and exit

  - `-t TITLE`, `--title TITLE` title of the experiment/HIT (default: None)

  

## assign_qualification.py

**Usage:** `assign_qualification.py [-h] [--value VALUE] QUALIFICATION [WORKER [WORKER ...]]`

Assign a qualification to a given worker ID/IDs. Useful for setting up

invitation-only makeup HITs.

#### Positional arguments

  - `QUALIFICATION`  qualification ID

  - `WORKER`         worker ID (default: None)

#### Optional arguments

  - `-h`, `--help`     show help message and exit

  - `--value VALUE`  qualification value (default: 1)

  

## create_qualification.py

**Usage:** `create_qualification.py [-h] NAME DESCRIPTION`

Create a new worker qualification. Useful in preparation for making

invitation-only makeup HITs.

#### Positional arguments

  - `NAME`         name of the new qualification. used to represent the

               qualification to workers

  - `DESCRIPTION`  long description for the qualification. this is displayed when

               a worker examines the qualification

#### Optional arguments

  - `-h`, `--help`   show help message and exit

## psiturk_batcher.py

**Usage:** `psiturk_batcher.py [-h] [-m MAX_ASSIGNMENTS] [-s SLEEP_TIME] n_assignments reward duration`

Emulate TurkPrime's HyperBatch feature to avoid accruing an extra 20% MTurk fee

for having more than 9 subjects / HIT. Based on Dave Eargle's `psiturk_batcher.sh` script.

Note: Before running this script, make sure that `launch_in_sandbox_mode =

false` in the psiturk config.txt so that it creates live HITs!

#### Usage

Place in the same directory as the experiment's `config.txt`.

#### Positional arguments

  - `n_assignments`         total number of assignments to post

  - `reward`                reward (in USD) for completing a HIT

  - `duration`              maximum amount of time (in hours) allowed to complete

                        the HIT

#### Optional arguments

  - `-h`, `--help`            show help message and exit

  - `-m MAX_ASSIGNMENTS`, `--max_assignments MAX_ASSIGNMENTS`

                        maximum number of assignments for any individual HIT

                        (default: 9)

  - `-s SLEEP_TIME`, `--sleep_time SLEEP_TIME`

                        time (in seconds) to sleep before posting a new batch

                        (default: 5)