Style guide

We try to conform to some common stylistic, writing, and study design guidelines to provide a familiar and friendly experience across Lookit studies.

General language usage

  1. Use casual language and terms where possible
    • “Baby” rather than “infant” (they mean the same thing, but “baby” is far more common in everyday speech; “infant” is more common in clinical/academic settings)
    • “Newborn” rather than “neonate”
    • Referring to the parent/guardian can be tricky since they may or may not be a “Mom” or “Dad” - “grown-up” works well in most cases
  2. Use singular “they” rather than “he/she” or “(s)he” or “the child.” (E.g., say “Hold your child so they can see the screen” rather than “Hold your child so he/she can see the screen.”) It’s common enough now that it’s more readable, plus it encompasses families who use pronouns other than he/she.
  3. A girl is a female child. If you’re talking about a female grown-up–e.g., a video is going to show your RA demonstrating a toy–that’s a woman. (You can consider words like “person,” “somebody”, or “friend” if that sounds awkward!)
  4. When asking questions about the child’s family or caregivers, be explicit about who you mean. It may feel more “inclusive” simply to say “family,” but for families dealing with multiple definitions it’s confusing to try to figure out what you really care about - especially since research is weird and who knows how something that feels irrelevant might not be!

Music selections

When choosing music or nursery rhymes for stimuli or background music, please confirm that it’s something that you would be comfortable sharing in a preschool classroom! In particular, a lot of US “children’s songs” turn out to have disturbingly racist histories. Even if the current lyrics are inoffensive, please don’t use music that is known for past use in minstrel shows, for instance, or where there are alternate historical lyrics that are offensive. Examples of songs not to use include “Oh Susanna,” “Camptown Races,” and “I’ve Been Working on the Railroad.” Here’s a good introduction to this topic if you’re interested in learning more!

Eligibility criteria

In general, keep eligibility criteria as broad as possible for a positive family experience. The “hard” eligibility criteria you list on Lookit will include your age range and any characteristics that are critical to the study design or that are necessary for the study to make any sense to the child/family. For instance, if you have a question specifically about bilingual vs. trilingual kids, go ahead and require n_languages >= 2. But if you just routinely exclude premature babies or kids with trisomies from analysis, you can do that after the fact if needed since the study experience won’t be affected.

Your eligibility criteria description should “translate” the minimum / maximum ages and any eligibility criteria expression into regular language, starting with “For…” - for example, “for 3-year-olds” or “for 8- through 12-month-olds who are exposed to least two languages at home.” You can also include hard criteria that are not possible to check for programmatically, for instance if your IRB requires that participants live in a certain country.

Specific guidance:

  1. If recorded audio in a particular language is a critical part of your study, note that the study is for speakers of that language. (You don’t need to separately exclude deaf children!)
  2. If you need kids to have typical hearing to participate, and that’s not otherwise obvious from the design or a language requirement, list that as “with typical hearing” or “without hearing loss” rather than “normal hearing.” Only do this if hearing is critical to the design and/or to family experience.
  3. Don’t require babies to be born at full term in your hard criteria unless the age range extends under 6 months, your study is specifically about effects of prematurity, or the study requires unusual time and effort from the family. Otherwise MANY studies exclude preemies and it’s frustrating for parents. Plus this pushes all of us towards putting some upper bounds on how much of a difference prematurity actually makes. Some enterprising student should ask everyone for data excluded due to prematurity eventually…

Study purpose

The study purpose is the most challenging piece of the study to write for most researchers (yes, including all that JSON!). The big questions to consider are:

  1. Is this an ACCURATE description of the question the study will answer? I.e. when we get the data, is it at least possible that we will be able to answer it? Often we see descriptions of a broad research program, instead of the particular question THIS study will address.

    Examples of questions your study does not answer because they are way too broad: “How do children learn language?,” “How do babies think about morality?,” “How does numerical cognition develop?”

  2. Does it explain why this question matters or why this question is interesting? This doesn’t need to be a “practical” reason like informing parenting or education practices - basic research matters! Here are a few common framings that do NOT explain why a question matters:

    • Not much is known about X.
    • X is an important skill for doing Y and Z, so we’re interested in whether babies can do X / how babies learn to do X.

    However, you don’t have to justify the entire field of cognitive science/psychology. E.g., if your question would distinguish between there being two separate systems for representing number that kids have to link, and there only being one, you’re set. Distinguishing between competing theories and determining whether things are innate are generally ok places to stop.

Examples coming soon of good “purpose” sections and before/after revisions!