Complete here means the full end-to-end sequence of all chromosomes in a single human cell line named CHM13. The typical human cell has 46 chromosomes, in 23 pairs (one from our mother, one from our father) named chromosome 1, chromosome 2, and so on. This CHM13 cell line is special is that each of its pairs is (nearly) identical. Each chromosome is a long string of A,C,G,T nucleotides. So, this complete genome is a full set of 23 sequences without any "not sure" positions or "gaps" in the sequence.
One common analogy is to consider the genome sequence (a.k.a. assembly) as a map. Since the initial publication of the human genome in the early 2000s, most regions of human DNA has been known in full resolution. Other portions, most prominently the repetitive centromeres that lie at the middle of chromosomes, have remained unmapped. It was known that they exist, approximately how big they were, and which types of sequences lay inside, but the full order of the sequence had never been determined for any human genome until this work.
You could consider the genome like the earth and the centromeres like a dense rainforest. Previously we had detailed maps of most of the earth, and we had mapped the boundaries of the rainforest and had satellite-level images (i.e. we knew they were full of plants). Now we have on-the-ground pictures with full detail.
Having a map of these sequences makes the accessible to study. One of the most valuable uses of the human genome is as a shared coordinate system used by scientists to compare different individuals and identify and name genetic variants that explain human traits. We lacked that coordinate system for a big chunk of the genome until now.
As you say, this paper reports the sequence of a single human cell line named CHM13. Each of us has a slightly different genome sequence (really two of them, one from each parent). Now when scientists sequence the genomes of more individuals, they can look at these regions that were previously ignored. Certainly understanding those regions will improve our understanding of human biology. Exactly how much will remain to be seen.
Survata co-founder here. To clarify, Survata is not a voluntary response sample. Voluntary samples often have a bias because the individuals who choose to respond are those with strong feelings on a topic. For our surveys, the primary incentive is access to premium content - and not a desire to express one's opinion on a topic. We aim to have a respondent pool that truly represents the population.
Good point. We had the same thought and did consider running a survey variant with the fixed 6 mo time frame. And we may just give it a try to see how the results are affected.
Even with the current wording, I find the SMS comparison useful. It demonstrates that people are willing to admit to sexting in the anonymous Survata survey format. I like your hypothesis about greater willingness to admit to "bad behavior" in the distant past than in the recent past. My intuition is that anonymity weakens that effect, but we'll have to measure to know for certain.
That's one of the first things that we (I'm a Survata co-founder) noted in seeing the data too.
I see a few possible explanations:
(1) A woman who sexts could have multiple sexting partners. In the extreme, you could have every man in the world sext with one woman, making the male sexting prevalence 100% and the female near 0.
(2) While we defined sexting as "sending or receiving", some respondents may interpret the question as primarily about sending. There could be a gender bias in the sending vs receiving of sexts.
(3) As you point out, the data is reported behavior and not observed behavior. Reported behavior often is a good proxy for observed behavior, but it is not perfect. And there are known to be effects where certain demographics answer questions dishonestly for conscious or unconscious reasons. Perhaps women are less willing to admit to sexting behavior.
Survata has a DIY survey creation tool, but we review and suggest wording changes to avoid biased questions. We also advise on how to arrange (and randomize) answer choices to allow us to calculate and compensate for answer biases like always clicking the first or last option.
Responses are gathered on surveywalls across the web, where visitors answer short surveys in exchange for free access to premium content (e.g. ebook or video).
Garry's explanation is a good one. The data for this survey was collected via surveywalls (example at [1]), which let visitors access premium content online for free in exchange for answering a few questions.
All respondents here have US IP addresses and self-report age in the 13-25yr range. We generally see honesty rates of 90% or higher to questions for which we can verify the answer (e.g. "Which OS are you currently using?" or "Who is the President of the US?").
Thanks again, Alex. I appreciate the comments and great ideas.
We do plan to add polling and survey design experts to our team (I have stats but not polling background). And an analytics platform is coming.
What do you mean by our survey "might take the user by surprise"? Do you mean that it might be surprising to the user to have a survey launch when clicking one of our links? We're testing different "teaser" text for the links. And hopefully we can make it clear that a survey will be coming when you click.
If you're willing to run a test survey, fill out the contact us form on our site, and we'll give you a discount code. Appreciate the feedback.
I've enjoyed the recent 5-hour ENERGY ad about how many doctors approve of their product. A whopping 73% of doctors recommend low calorie energy products... when measuring the percent of those who recommend energy products.
Thanks for the interest, and sorry for the delayed response.
We certainly hope that the surveys are not annoying. We know that some people will prefer to pay, but others (like me) would rather take a few seconds to complete a survey than pay. Our hope is that the surveywalls will let users access content that otherwise would have been beyond their reach. And at the same time, quality content publishers will be able to make money off their work.
As to the researcher side, we agree with you that nothing beats revealed user behavior when optimizing web sites or apps. But it's not cheap or even possible to A/B test in other situations.
Suppose that you're a restaurant owner and want a new sign for your building. It's not practical to purchase two signs and see how alternating the signs affects business on different days. And it's also not in budget to spend $10k or more on a traditional market research survey.
Or if you're a politician, you can not wait until voting day to see which of your various ad campaigns worked in different districts. You need proxy measurements.
Companies already ask these types of questions using traditional approaches like panels and phone polling. We provide a cheaper way for them to do it online. All approaches have built in biases, and we're working to account for the biases and quality issues in online polls. As you point out, we'll have to do that to succeed.
While we hope to work with big publishers too, niche blogs are an awesome place to start. As you say, they provide a self aggregated group of individuals with a common interest. They're perfect for polling.
Thanks for the interest. Google Consumer Surveys does have a similar model to ours. There a few things that differentiate us now:
1. Our minimum spend is only $10 vs. $100 for Google; lower minimum brings in a group of people who want to get a "quick read" on an issue (e.g. ask 100 people an opinion on a new logo design or tagline)
2. Starting soon, we'll be offering advanced behavioral targeting. For example, you could select a target audience of active young mothers, Honda car owners, or online shoppers.
3. We're allowing multi-question surveys (up to 4 questions). Each respondent answers all questions in your survey, which enables cross-tabulation of responses (e.g. looking at how respondents who answered "Yes" to question 1 respond to question 3). Google is currently focused on 1-2 questions at a time.
In the coming months, we hope to differentiate ourselves in other areas, in particular data quality.
We agree that maintaining data quality is the biggest challenge and really the crux of this business. We have initial filters that look for incorrect answers to "checker" questions (e.g. What browser are you using? In which time zone do you live?), an outlier response speed (too slow or too fast), etc. And we're working on a more sophisticated system to really address this.
Data quality with initial partners has been good. We'll be working hard to keep it that way.
I agree that the lightbox experience is common, but I view the surveys as fundamentally different than advertisements. Part of our idea is that your attention is more valuable than a banner/video ad. If I had 30 seconds of time from a smart HN reader, I'd rather ask for advice than show a commercial. Hopefully getting at that valuable opinion will allow you to get better free content online.
And, just to be clear, we're keeping the surveys fully anonymous; so we will not collect or tie to an e-mail address or other personally identifiable information.
See https://github.com/marbl/CHM13 and https://www.science.org/doi/10.1126/science.abj6987.