S**t My Undergrads Say

I’m very fond of “ranking problems” that ask students to order a series of compounds in some way and provide an explanation. One of my all-time favorites is the famous “rank these acids from most to least acidic…” problem, which might include compounds like the following.

Rank the compounds shown from most to least acidic. Explain your response.

Rank the compounds shown from most to least acidic. Explain your response.

To really understand a problem like this deeply, the student has to be able to connect the structures provided with either physical/chemical properties or theoretical constructs (such as electron-donating and electron-withdrawing groups). Structure-property relationships are at the root of the appropriate thinking here. Unfortunately, as was pointed out by a recent paper that caught my attention in J. Res. Sci. Teach., students often struggle with structure-property relationships. Without experience under one’s belt, the incredible utility packed into a Lewis structure can be lost on students. It’s staggering, really—what other devices in science approach the information density of a chemical structure?!

I wanted to comment on this paper because it resonated with me. The authors explored the prior knowledge and heuristics that students used to rank the thermodynamic favorability of a series of chemical reactions. The work is a small piece of a larger movement that explores how chemists both novice and expert make judgments and decisions when solving problems.

I imagine that work in this field gets messy and discouraging very quickly. Novices are inclined to rely on unconvincing, arbitrary, or otherwise spurious assumptions. Is it worth drawing attention to these bogus ideas that guide students’ thought processes? That remains an open question, but I’d say that to conquer a problem, one must first hold it up to the light. As painful as it may be to learn just how little instruction affects students’ intuitive assumptions, “the first step is admitting you have a problem.” Chemistry experts, by the way, aren’t immune—just as many experts have trouble verbalizing their thought processes when navigating by feel!

In the paper, Maeyer and Talanquer studied problems related to acid-base chemistry, elemental synthesis, and combustion. They asked students to order sets of three related reactions from most favorable to least favorable in a thermodynamic sense. Qualitative interviews of a small set of students (33) supplemented large-scale survey data. The results were typical: a small number of students (less than 35%, generally) actively verbalized what the authors called valid chemical assumptions while rationalizing their orderings. Many more employed either intuitive assumptions that had at their core ideas completely unrelated to chemistry (“bigger things are harder to make than smaller things”) or spurious chemical assumptions with no basis in chemical theory but employing concepts from chemical theory (“electronegative atoms are more reactive”; bogus periodic trends).

With fascinating results, the authors also explored the heuristics students used to come to a decision. A majority of students focused on a single factor that they presumed to be the essence of the difference between the reactions—the authors called this the one-reason decision-making (ORDM) heuristic. The recognition heuristic played an important role, as many students stuck at the top or bottom of their lists reactions that involved familiar species (“Ooh, I’ve seen NaCl before, so its precipitation from water must be favorable”). Some students cited artificial correlations between variables, a strategy the authors called more A-more B (“Na to Ca to Al goes from +1 to +3, so I’m going to put them in this order”). Finally, the authors identified a representativeness heuristic that involved grouping items based on explicit similarities (“Cl and I in HCl and HI have the same charge, so I’m putting them next to each other in the list. S in H2S has different charge, so I’m putting it at the bottom of the list.”).

Painful to read, isn’t it? Many of the students generated correct orderings, but for all kinds of bogus reasons. Rather like buying your girlfriend flowers on Valentine’s Day because of a fight the night before…not exactly playing Casanova there! Of course, not all of the assumptions and reasoning strategies are bad. For example, ORDM reflects an analytical approach that teases out all the important differences between the reactions—the trick is weighing the differences against one another carefully instead of betting it all on a single factor.

Maeyer and Talanquer harbor no illusions concerning the generality of their work. Many times in their paper, as is typical of qualitative research articles in chemical education, they remark that the work used a small sample and lacks generality without further study. I think the work is still valuable for at least two reasons: (1) their framework of assumptions/heuristics provides a useful lens for diagnosing problems in the thought processes of students; and (2) it invites teachers young and old to take off their rose-colored glasses and reconsider how they teach. I don’t know about you, but I’d rather read about students’ issues with chemical reasoning before encountering it in my own classroom! This paper is one in a long line of studies that paint a descriptive picture of students’ chemical reasoning processes, and many more are to come. That’s “descriptive” as opposed to “prescriptive”: what “is” rather than what “ought to be” (although interestingly, Maeyer and Talanquer devote half a page to what ought to be in the form of the “target reasoning process”).

There are interesting parallels to ethics and philosophy here: should chemistry education researchers be focusing on what is rather than what ought to be? Why waste time on the way things are when we have so much to say to students about how they ought to think? Clearly, that “so much to say” doesn’t have as great an impact as we think it does—studies like this one demonstrate that convincingly. But will studying the way things are help us advance teaching and learning in a lasting way? I’m not sure. What do you think?

Here’s the full paper.



  1. I was just lecturing on this last week! I was going over phenol acidity using the same kinds of EDG/EWG in the ortho position… But the students struggle with it, and I feel partially because I can’t find the right words or analogies to clearly convey the ‘why’ behind the concept. How do you lecture this to your students to help them understand how modulating the electron density of the molecule modulates the acidity of certain protons?


    1. Az,

      One thing that can be useful is to note that EDGs and EWGs fit certain structural patterns. All donating groups have a lone pair on the atom bound to the π system, for example (not foolproof because of imines, hindered amines, etc., but you get the idea). All withdrawing groups, with the exception of unsaturated atoms (boron, carbocations), fit the pattern -X=Y, where Y is more electronegative than X (could be a triple bond, of course, or even a single bond if you want to lump in inductive effects). While introducing these patterns, I encourage students to be playful/creative with resonance…a Lewis structure is not so much an immutable truth as a useful starting point. I bet if you drew para-methoxyphenol as the O+/C- resonance form, students would jump on its relatively low acidity right away! 🙂 Got to somehow arm students with the conceptual tools and confidence to be creative…

      One of the tough things about that topic, I think, is that much of it is empirically derived from linear free energy relationships, then rationalized after the fact. It makes predictions tough. For example, just a few months ago I learned that cyano is typically more withdrawing than carbonyl. Here two factors pull in opposite directions: hybridization for the cyano, electronegativity for the carbonyl. Most students jump all over the electronegativity difference, but neglect the (critical, as it turns out) difference in hybridization.


      1. Yup, I use the Z: directly attached vs X=Y directly attached to generally depict EDG vs EWG, too. They’re typically ok with understanding that a group is EDG or EWG. They’re even typically ok with why EDG are typically o/p directors and EWG are typically m directors… It’s just this concept of modulating acidity I haven’t found a good teaching method for.

        Last week I used the ‘molecular electron density is like a blanket’ word picture, where some groups pull the electron density blanket towards themselves (and away from other things like acidic protons) and some push the blanket away from themselves (and toward other things like acidic protons). I think they accepted that, and we drew some resonance structures and all…

        I think the stumble was in the number of conceptual steps needed to get from EDG to more acidic. Plus it’s almost like you have to approach it from both ends. Acidic protons are a kind of electrophile. Electrophiles lack electron density. Anything to decrease the electron density further makes the proton more acidic and v/v. At the same time, this group over here is EDG. It pushes electron density towards the acidic proton. This builds up electron density near the acidic proton. If acidic protons lack electron density, and this group is causing electron density to build up near the acidic proton, then the proton must be relatively less acidic.

        Couple that with not all of the resonance structures make it all the way to the acidic proton (or at least its heteroatom…). When the negative charge stops on the ipso carbon of the group with the acidic proton, the concepts are even farther removed from each other…

        I appreciate what you said about the concept being rationalized after the fact. That makes me feel better 🙂

      2. ok… so this just came to me literally right now. I’ll work out my thoughts here in this comment and we’ll see where they go.

        So you’re sitting at the high roller’s poker table in a big fancy casino with your tux on and everything. Your girlfriend is hanging on to your shoulder because her man looks hot sitting at the poker table and all. If you’re doing well and the poker chips keep piling up in front of you, then your girlfriend is much more likely to stay attached to your shoulder 🙂 but if you’re doing poorly and your chips keep diminishing and ending up somewhere else around the table… your girlfriend is more inclined to so see what that guy over there is doing…

        or something like that. I’ll work on that and try it out with my class later this week or something 🙂

  2. So I tried out my poker table analogy last week, and put a question on the exam yesterday (benzoic acid vs meta-methoxybenzoic acid vs para-methoxybenzoic acid. Sadly, the analogy did not transfer well. Too many students relied on induction as the primary rationale for modulating acidity, thus got the trend backward. Only one mentioned the poker table analogy by name, and this person had the best score on this question – tho not perfect.

    The analogy needs to be reworked out of a second person narrative and into a third person narrative. Making the character sitting at the table ‘you’ puts an artificial point of focus, when the students should really be focusing on the girl. I’ll keep working at it 🙂


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s