Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.expectedparrot.com/llms.txt

Use this file to discover all available pages before exploring further.

Note:This page provides information about EDSL code methods for launching surveys, gathering human responses, and analyzing results from your workspace. For more information about building surveys interactively, see the Survey Builder page. For more information about launching studies with Prolific participants, see the Prolific studies page.

​
How it works

1
Create Surveys with desired types of Questions, and design AI Agents to answer them. Select Language Models to generate the responses.
2
Run your survey with agents and language models, generating a formatted dataset of Results.
3
Generate a web version of your survey with the humanize method, which provides a link for human respondents and a project page at your account where you can access responses interactively.
4
Share the web survey with human respondents. (See the Prolific studies page for information about launching studies with Prolific participants.)
5
Collect the responses using Expected Parrot methods to gather the responses in a Results object.
6
Analyze your LLM and human results together.

​
Example

Code and results for the example below are also accessible at a downloadable notebook at Expected Parrot. 
# Import modules from EDSL
from edsl import (
    QuestionYesNo,
    QuestionNumerical,
    QuestionLinearScale,
    Survey,
    Agent,
    Model,
    Coop
)

# Create a survey with different question types
q1 = QuestionYesNo(
    question_name="drive",
    question_text="Do you drive?"
)

q2 = QuestionNumerical(
    question_name="count",
    question_text="How many vehicles do you currently own or lease?",
)

q3 = QuestionLinearScale(
    question_name="enjoy",
    question_text="On a scale from 1 to 10, how much do you enjoy driving?",
    question_options=[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
    option_labels={1: "Hate it", 10: "Love it"},
)

# Create a survey with the questions
survey = Survey(questions=[q1, q2, q3])

# Create an AI agent to respond to the survey
agent = Agent(
    traits={
        "persona": "You are a middle-aged mom working on a software startup.",
        "location": "Massachusetts",
    }
)

# Select a language model to generate the responses
model = Model("gemini-1.5-pro", service_name="google")

# Run the survey with the AI agent and model
results = survey.by(agent).by(model).run()

# Generate a web-based version of the survey for human respondents
web_survey_info = survey.humanize()

# Create an Expected Parrot client instance
coop = Coop()

# Get human responses from Expected Parrot
human_results = coop.get_project_human_responses(web_survey_info["uuid"])

# Combine results (you can add Results objects for the same survey)
combined_results = results + human_results
We are continually adding features for launching hybrid LLM and human surveys, so check back for updates! If you are interested in testing new features please reach out at anytime for credits and access.