Problems with Abstracts

A comment from M on my post about writing abstracts raises a problem that particularly concerns students and trainees, in response to my rejoinder to avoid the “results will be discussed” (slightly edited for space):

I… am more often than not faced with the situation of having to submit abstracts on work that is not yet done. … I often don’t know what the results will be, or if we will even get results at all, let alone having any conclusions. In these cases, I typically lay out the motivation for the work, give some vague description of what we plan to do, and then resort to “results will be discussed.”

The honest, honorable and without-question-correct thing to do is not write the abstract if you don’t know the results. Besides the ethical/moral justification for this, there is the problem that if you don’t have results you will tend to look foolish or worse, like a bad scientist, or even worse, like a lousy student pretending to be a scientist, as you try to piece together the shards of an experiment to present. I’ve been to too many of those talks, and no matter how hard I try to give the student a break, I’m frustrated because I went to the talk for the results.

That all said, the pace of work (now, as well as then, really and truly), makes it very difficult to do that. Even back in the day (when btw, stipends for grad students were ~5-6K$ in today’s dollars), no one who ended up being a 7th year PhD student intended to do that at the start. Everyone was eager to go to meetings, and present their results. Its how you get known, how you make contacts and often how you get pre-interviews for a job. So what is there to do?

Firstly, if you have absolutely no preliminary data, nothing, nada, rethink putting an abstract in. Yes, I know that you know that you will have glorious data in 6 months. But the best laid plans of men, mice, and other model organisms often go astray. [See aside below about meeting attendance]. The potential cost of not having data can be very high indeed. Your reputation as a scientist is important, and nothing like a couple of “results will be discussed” abstracts to make people avoid your talk or poster in the future.

Secondly, if you do have some preliminary data, it is valid and acceptable to say something along the lines of “Preliminary results indicate that X, Y, Z“. Then in the conclusion/discussion you can (depending on level of boldness, your mentor, co-auths, etc) include something like:

(1) If further data bear out these results, then blah, blah, blah.


(2) These results (implicitly the preliminary stuff) suggest that blah, blah, blah. 

or even

(3) If we find that G&H (your best prediction) are true, together with these results, it would imply that blah, blah, blah.

One of the other things to be careful of is being perceived as presenting the same thing year after year. The response you eceive may not be right or fair, but it happens: you will be ignored.  This is as true for a grad student presents every year on the marginal increments of their work as it is for deadish wood who retwirls their data and presents the same stuff with a new spin (there is a famous case in a field that I occasionally go to where the same single case has been re-interpreted several times and presented 3-4 years in a row. I missed nothing by skipping that meeting several years in a row). Deadish wood may not be overly concerned with being ignored, but a student or postdoc on the job market does not have that luxury. What happens is people at meetings (and at good meetings there is way too much to do and see and too many people to talk to) is that they make choices. They make choices on what to see based on the abstract. Given a choice between “results will be discussed” and a set of very compelling results that speak to my next grant, I’ll pick the latter.

The aside: you don’t have to present. You can go to meetings to absorb. Sometimes that is a good thing. If this is going to be your chosen career, investing in a trip to a meeting, and doing it cheap can be incredibly valuable. Pick the meeting wisely, share a room, and eat breakfast from the candy available in the exhibits (I admit to doing this and being dreadfully ill as a result).


One thought on “Problems with Abstracts

  1. All good points. A few thoughts… (sorry, this turned out kind of long)

    Some of the attitude toward this issue may be field-dependent (health/bio vs other sciences), because I feel like the vague abstract approach is relatively accepted around these parts (but perhaps I need to inquire with more people). But there is also some situational context that could make a difference.

    Typically when I (or more often a student) submits a vague abstract on new science, they are planning to present a poster and not a talk, so the risk of them wasting peoples’ time is minimal. On the more rare occurrence that a student (or myself) submits a vague abstract for a talk, the talk would likely be at a big society meeting in one of the obscure sessions, likely clustered with a number of other grad student talks, with probably 5-15 people in the audience. I think in both of these cases, the ‘experience’ of preparing and presenting a poster or talk is worth more than any negative ethical implications (but nothing untrue is ever implied, so I guess I don’t feel that this is particularly ethically questionable). Perhaps the issue is that there are minimal small-community field-dependent meetings for us to go to in the first place, and it is rare that I would find myself in a situation where the audience would be drooling over learning about my new results. When this type of opportunity does present itself, I would strongly hesitate to send a student (or myself) to a well-attended symposium/session where people actually care about our work with an unsubstantial talk and/or unsubstantial results. That said, there are borderline cases where the work is partially done and we are confident that we are getting results, but just how far we will get and how awesome the results will be in 6-9 months is somewhat inconclusive.

    I also take care to not get people’s hopes up – it should be clear from the abstract that our results are unknown and it may or may not be worth their time to attend the talk. They will know, however, what material we’re studying, what property we’re investigating, what method (or at least broad set of methods) we’re using, etc, so they will know if they are interested in the subject matter. We wouldn’t submit something that said “we will present our incredible results demonstrating x as a new example of y.” I believe some of the worst offenders are the abstracts that give extremely broad titles and descriptions of the work (e.g., unifying theory x in technology y) and then end up being some ultra-specialized talk on something you’re completely uninterested in than has nothing to do with theory unification.

    But really, I should say that I’m jaded when it comes to conferences/talks overall. Because I’m more often than not attending meetings with a huge variety of speakers and attendees in very different fields (or at least sub-fields), I think the most important thing is to give a good talk, clearly present the motivation for your work, dumb the results and story down enough such that the non-specialist can understand the takeaway and the importance, etc. This is so rarely done. Yes – in an audience of specialists, one might delve into the details – but this happens so rarely that it’s not the norm. It’s more important to me to hear a good talk that I understand and can be interested in and can learn from than one reporting some amazing new cutting edge result because this just usually isn’t why I’m at a meeting in the first place. My hope when I give a talk is that people leave thinking “Wow – M gave a really good talk and is doing some respectable science. It would be worth my time to hear a talk by this person again, read their paper, speak with them further, etc.” more so than being blown away by my results (though this would be a nice bonus). The only people I’m trying to blow away by my results right now are my program managers, ha.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s