Analytics and Advising:
Seeing signal in the noise

In the Internet age, the problem often isn’t a lack of information—it’s figuring out how to sift through it to understand the world.

If that sounds like a challenge you face, we can help.

Using rigorous, cutting-edge research methods, we can develop and implement a quantitative research strategy at any phase of the research process: question formulation, data collection and processing, analysis, and presentation of results. Whether you’re looking for someone to develop and manage a full project, or a specialist to play a role on an existing team, we can help.

Our fullest process usually looks like this:

Idea Generation and Assessment of Needs and Resources

Not all questions can be addressed by analytics research, but sometimes a quantitative approach can be applied in surprising places. We would love to meet you in an initial consultation to understand what information you already have, and what you might need.

We will “mind meld” with you to assess how to formulate your questions as tractable for an analytics approach involving the tools of hypothesis testing and data science, and whether we can use your existing data or need to create a plan for gathering more.

Proposal and Design

After our initial consultation, we will develop a research plan involving one or more tools from our analytics toolbox. This plan will include a schedule and budget for implementation, as well as clear statements of what we can learn from this plan and why this approach is appropriate for the questions you want answered.

Implementation

Then we do the legwork! We conduct original research specifically tailored to your need. We’ll be in continual touch during the research process as challenges or unexpected findings arise. We use academic-level analysis tools to generate results you can be confident in (see the toolbox below).

Presentation, Feedback, and Future Steps

Finally, we present our findings to you or your preferred audience in a clear, visually appealing format such as slides or a brief report. We highlight the key takeaways and advise on next steps and how this project can fit into your longer-term research agenda.

If this sounds like a good fit for your needs, please reach out at tko@tkoresearch.org

We’ll match the right tools to your project.

 

Research frameworks

Different research questions mean different research approaches. A huge amount of social science training is required to know what kinds of approaches can answer which kinds of questions. Fortunately, we’ve invested in this training and gained practical experience implementing these studies to know which tack to take.

Some approaches we use include:

  • Program evaluation

  • Analysis using existing data

  • Descriptive survey analysis

  • Survey experiments

  • Field experiments

  • Random controlled trials

  • Spatial analysis

  • Causal pathway analysis

  • Institutional policy analysis

  • Decision and game theory

  • Power mapping

 

Statistical tools

Having the data isn’t much use if you don’t crunch the numbers right. We use the appropriate tools to make sure the signal we detect are reliable and can generate confident, actionable insights.

Some of our statistical tools include:

  • Cross-tabulations

  • Data visualization

  • T-tests of experimental group differences

  • Multivariate regression

  • Matching techniques for causal inference

  • Difference-in-difference estimation

  • Text-as-data content analysis

  • Network analysis

  • Factor/Principal components

  • Spatial econometrics

  • Multilevel modeling

  • Survey weighting and adjustment

  • Time series and panel models