Designing new types of problems using peer-reviewed papers

In the 21st century teachers, including university teachers, have to help students develop higher-level thinking skills, conceptual understanding, and problem-solving strategies similar to those used by experts. Previous research shows that solving traditional physics problems, which usually have one right answer and provide all the given and relevant assumptions, do not help students develop those skills. Thus, new types of problems have to be designed. We show two examples of how new types of problems can be designed using peer-reviewed research papers as the basis and share the experience with the evolution of those problems. In addition, we show that the students included in the research appreciate the structure and the context of the problems and that the faculty colleagues who are using traditional teaching approach identify several skills and competencies which these problems develop, and which cannot be developed by solving traditional problems.


Introduction
Traditional end-of-chapter problems in introductory physics textbooks normally ask for a specific numerical value, while giving all necessary data and making all simplifying assumptions that are needed to find the solution. Students often solve this type of problems by searching for an equation in which they insert given numbers to get an answer (so-called plug-and-chug problem-solving approach [1]). Research showed that while solving a large number of traditional problems may help students improve their mathematical skills, they do not improve conceptual understanding. Research found that common difficulties in understanding basic concepts remain even after the students have solved over 1000 traditional problems [2]. Another research suggests that the students who have good conceptual understanding are more likely to detect their own comprehension failures while the students who do not have good conceptual understanding will know this only when they cannot follow mathematical manipulation [3]. In addition, today we have irrefutable evidence that students learn better through interactive engagement methods than through traditional transmissionmode methods [4][5][6]. Finally, new recommendations to educators, driven by the demands of the labor market [7], call for helping students develop higher-level thinking skills, conceptual understanding, and problem-solving strategies similar to those used by experts, and using active learning approaches. Consequently, one of the new tasks of physics educators is to design and use new types of problems that help students develop abilities such as those described above. Yet, designing new types of problems is much more difficult than designing traditional problems, which is one of the reasons why new types of problems are rarely found in textbooks.
The goal of this paper is to share with the readers our experience in designing new types of problems using existing research papers from physics peer-refereed journals as a starting point. We describe the design of problems based on two physics research papers and share some experience in piloting these problems with students. We discuss the benefits of new types of problems as seen by students and faculty and provide suggestions for those who wish to engage in the design of such problems for their students. The new types of problems that we present are suitable for any university-level introductory physics course.
In the next section, we define the term new types of problems and describe some previously published new types of problems that have the potential to be successful in helping students develop the competencies described above. We then present two such problems that we designed using peer-reviewed research papers as well as some responses from the students and from the faculty. Next, we discuss what could students learn from these types of problems that they cannot learn from traditional problems and end with offering some advice to those who want to design new types of problems using peer-reviewed papers.

New types of problems
To our knowledge, there is no common agreement about the definition of the term 'new types of problems'. In this paper, we will use this term for the problems, that fulfill two conditions: (1) they help students develop conceptual understanding and (2) go beyond the problems that have exactly the givens and all relevant assumptions that are needed and ask for a specific quantity. Below we describe a selected number of new types of problems that are relevant to our paper. Interested readers can find an extended list of new types of problems in [8,9] and numerous examples of those problems in a textbook [10].

Ranking tasks
The problem usually starts by describing several contextually similar situations that differ in the value of one or more physical quantities (for example, liquids of different densities). We ask the students to rank the situations in ascending or descending order, according to some other physical quantity (buoyant force exerted by the liquid on a given object). We can also ask them to explain how they came up with their ranking. Solving problems of this type helps the students develop the ability to compare different situations and to visualize the situations. Problems of this type allow the students to demonstrate various levels of achievement and therefore give every student opportunity to feel successful. Ranking task problems are described in the literature [8,11] and are regularly used in national exams such as AP exams [12].

Choose answer and explanation
The problem typically starts by describing a situation and asking the students to identify relationships among the physical quantities or to predict what will happen next. The students are given several options from which they must choose the one that combines a correct answer and the best matching explanation. When designing such problems, it is important to pair the correct answer with at least two different explanations that differentiate between the students who have deep conceptual understanding and those who have only a superficial understanding of the situation. Solving problems of this type helps the students develop the ability to recognize the correct cause-effect relationship in addition to finding the correct answer and to develop argumentation skills. Problems of this type are regularly given on AP exams [12].

Evaluate reasoning or solution
The task typically starts by describing a problem that was given to one or more (imaginary) people and continues by describing the solutions provided by these people (given either in words, graphs, diagrams, equations or all of the above). The students must critically evaluate the reasoning and the suggested solutions. The students are asked to recognize productive ideas and to differentiate them from unproductive ideas. Solving problems of this type helps the students develop the ability to focus on other people's reasoning and to recognize productive ideas even when they are embedded in incorrect answers. Such problems allow the students of different levels of achievement to feel successful. Problems of this type are relatively new [8].

Make judgment based on data
The problem typically starts by describing a phenomenon or an outcome of an experiment including relevant data or some other forms of evidence. The students have to make a judgment about one or more hypotheses, based on the data or other evidence, sometimes taking uncertainties into account. The problems of this type help students learn how to reject a hypothesis and how to analyze data. Problems of this type are relatively new [8].

Linearization problems
The problem can start as a traditional problem where the students must find unknown quantities based on given data. First, the students have to write an equation that describes the problem situation. Then they have to rearrange the equation to obtain a linear function (note that the independent and the dependent variables in this function can be any function of data given in the problem). The students then draw the graph, plot the best-fit line, and determine the unknown quantities using the best-fit line. These problems help the students combine knowledge of physics, the ability to 'read and write' with graphs, the ability to manipulate equations, and the ability to recognize linear dependence in non-standard situations. The problems of this type have been used in Modeling Instruction approach [13] for two decades and are sometimes given on AP exams [12]. The students have to list as many quantities as they can that can be determined based on the data given in the problem or to tell everything that they can about the physical attributes of the objects that appear in the text or the relations between them. Normally, the students are required to determine the values for only a few of the quantities that they identify. Because the number of quantities that the students need to list is not specified, these problems also allow all students to feel successful. Problems of this type are relatively new [8].

Jeopardy problems
You can think of these problems as 'backward' problems. The students are given a representation of a solution (equation, graph, force diagram K) and they have to propose the problem statement [14]. If the solution is given in the form of an equation, they need to understand the meaning of the quantities and their units. Problems of this type allow students to be creative and allow for various levels of achievement. Many problems of this type can be found in [10].

Problem I: Electrostatic force between two charged metal spheres
The original paper that served us as a basis for designing the problem for the students describes a precise calculation of the electrostatic force between charged spheres including induction effects. The paper was published in 1990 in the American Journal of Physics [15]. The author of this theoretical paper used the method of images to compute precisely the magnitude of the electrostatic force exerted on each other by two charged metal spheres (equal spheres, equal charge magnitudes) and to compare it with the force that the same two spheres would exert on each other if the charge distribution were spherically symmetric. The calculation shows that as the distance between the metal spheres decreases, the force exerted by equally charged spheres on each other is weaker and the force produced by oppositely charged spheres is larger than the force produced by the same spheres with a spherically symmetric charge distribution. The author of the paper presented the results of the computation as a graph (see figure 1), which served us as the main source of data in designing our new problem.
We decided to use the graph from the original paper to design a new problem that requires students to make judgment based on data (see section 2). In our problem, the data are presented as (hypothetical) experimental data. We decided to modify the representation of the data so that Coulomb's law is represented as a linear function (see Linearization type of problem in section 2). We also gave values for the force in newtons, using suitable realistic charges. The complete problem including new graphs is shown below. We recommend giving this problem to students at the end of the unit on electrostatics.

Type of problem: Make judgement based on data
In one laboratory, the scientists measured how the magnitude of the force F the two charged identical spheres exert on each other depends on the distance d between the spheres (see sketch below). They performed four experiments, which are presented in the following table: Radii of the spheres Charge on the left sphere Charge on the right sphere The measurement results are presented in the graph below (note that the same measurement results were obtained in two experiments).
a. Based on the data in the table and graphs, decide which measurement (A, B, C) belongs to which experiment (1, 2, 3, 4). b. Explain your choices by drawing charge diagrams.
Hints: Compare distances d with the size of the spheres in the experiments. Remember that spheres are conductive and that the graphs show magnitudes of the forces (absolute values) exerted by one sphere on the other sphere.
We gave the problem to eight master students of our Physics Education Program. All students previously completed a 3 year physics major program (first-cycle degree). We chose those students because they were familiar with the variety of new types of problems, which are a part of the teaching-learning approach called ISLE that we use in our Physics Education Program (ISLE is an acronym for Investigative Science Learning Environment [16]). The students solved the problem in groups of four using small whiteboards to present their solutions. We monitored group work and occasionally helped students by posing additional questions. Our observations of student work and their responses are summarized below: • All groups needed some time to recognize that the horizontal axis of the graph represents d 1 2 / values. Once they recognized this, they all quickly realized that graph B corresponds to the familiar case of Coulomb's force produced by two point-like charges or two spheres with spherically symmetric charge distribution. Next, all groups realized that such charge distribution can be obtained in experiments 3 and 4, because the distance between the metal spheres d in these cases is much larger than their radius through the whole experiment (some groups needed to be prompted to visualize the distance between the spheres compared to their size in different experiments).
• When discussing graphs A and C, all groups came up with the idea that the deviation from the linear dependence (graph B) is the result of the influence of charges on one sphere on the charges on the other sphere. Realizing that in experiments 1 and 2 the smallest distance between the spheres is only slightly larger than the diameter of the spheres helped them conclude that graphs A and C correspond to experiments 1 and 2.
Visualizing the excess charges (positive or negative) as mobile charges and taking into  account basic rules for interactions between the charges, all groups concluded that graph A corresponds to experiment 1 and graph C corresponds to experiment 2. The charge diagrams that students drew were similar to those in figure 2(a) (experiment 1, graph A) and figure 2(b) (experiment 2, graph C). In some groups, the idea of effective distance between the charged parts of the spheres emerged. Or as one of the students described (this and all other student comments are translated to English): 'First I realized that there are freely moving electrons, because it is a conductor, which means that the charges on the sphere can (and do) redistribute, thus, the centers of charges either move apart (for 1 and 3 since the charges are of the same type) or come closer together (for 2 and 4)'.
• When asked to explain the outcome of experiments 1 and 2 using the knowledge that in the conductors, negative charges (electrons) are mobile and positive charges (ions) are fixed, all groups struggled. The difficulties that students had mostly resulted from their attempt to take into account interactions of all charges (on both spheres) when deciding where negative charges will move when we bring the spheres close to each other. It proved to be helpful to discuss with the students two issues. First, the direction of the net force exerted on the electrons on one sphere (when we bring the spheres close to each other) is determined by the sign of the excess charge on the other sphere. Second, if the electrons move from region 1 that is neutral to region 2 which is positively charged, then region 1 becomes positively charged and region 2 becomes neutral. Representing this process as shown in figure 3 helped students realize that this is equivalent to having only a positive charge and moving it from region 2 to region 1. • In reflection, all students expressed positive opinion about the problem. Many said that they had found the problem insightful, particularly the part where they had to draw charge diagrams, taking into account that the electrons were the only mobile charges. Or as one of the students described: 'The problem attracted me because I could rely on my mental image of the experiments. K I think, the problem was most useful for me because it made me deeply aware that in the conductors the mobile charges are only electrons.' Testing the problem with our students in the Physics Education Program which are used to the ISLE teaching approach showed that the students found the problem useful. However, many students are taught in a traditional way. This raises a question: what do traditional teachers think about this type of problems? Do they also see them as useful? To get an answer to this question, we gave the problem to seven faculty colleagues. They all have a PhD in physics and several years of experience in lecturing or conducting problem-solving sessions in introductory physics for physics majors. All of them use traditional methods of instruction, including traditional physics problems. We asked them to solve the problems and to answer a short questionnaire. In the questionnaire we first posed two open questions, (i) 'In your opinion, what knowledge/skills in addition to the ones developed by traditional problems will students develop when solving this problem?' and (ii) 'Please, write down any other thoughts regarding physics, teaching or learning that come to your mind in relation to this problem.' In addition, we asked them to choose from the list that we had provided all those competencies and skills that they think a student would need to solve the problem.
In open reflections on the problem, the faculty listed several physics concepts (such as electric induction, electrostatics of conductors and mobility of charge carriers) and skills (such as checking limiting cases and reading graphs). Three of them emphasized the ability to solve qualitative problems, though two of them associated this ability with intuition (in their own words, 'The ability to think about the phenomena intuitively, not via equations. K' or 'Intuition for qualitative behavior of the solution of problem with an unknown analytical solution K'). Even though all the faculty who participated in the survey use a traditional approach in their classes, they all identified the following three abilities that are typically not addressed by solving traditional problems but are relevant to solving the presented electrostatics problem: the ability to relate graphs to the corresponding experiments, the ability to qualitatively describe/explain the processes, and the ability to propose different explanations for the observed patterns. All in all, the faculty found the electrostatics problem very challenging and far beyond the traditional problems in electrostatics. However, they also found it instructive and representing an example of how two effects which are usually considered separately can be combined and discussed in a qualitative manner ('Difficult and instructive problem. K' or 'I have to think of a problem like this . This is not easy, because such problems are usually related to what we would call in jargon 'second approximation', when there is not enough to consider the primary phenomenon, but you have to add the phenomena we usually neglect. K').

Problem II: Torricelli's law and the time to drain a tank of water
Another original paper that served us as a basis for designing a problem for the students describes a derivation of the mathematical model for the draining of water from a reservoir and its exact solution. The paper was published in 2021 in the European Journal of Physics [17]. The author revisited the drainage problem both theoretically and experimentally. The model assumes an incompressible fluid and is based on Torricelli's law, which states that the speed v of a jet of water leaving a reservoir a distance h below the water surface is = v gh 2 .
The model neglects the friction effects and the moving of the surface of water. The author derives a prediction for the time to drain for the cylindrical tanks with various cross sections with a focus on a tank with a shape determined by the equation = h cr 4 where r is the radius of the tank at height h and c is a constant. Tanks shaped like this have a special property that when drained, the water level decreases with a constant speed, which makes them useful as water clocks. The author continues by comparing the drainage times predicted by the model with those obtained experimentally and reports large differences, well above the experimental uncertainties. In the rest of the paper, the author describes the improved model that gives results that are in better agreement with the experiment.
We have used the expressions for the shape of the tank and the time to drain it, the given characteristics of the experimental setup, the measured data of the time to drain, and the reasoning about the validity of the assumptions used in the derivation, to design a new problem. The problem requires the students to (i) make judgments about the model based on data and (ii) evaluate the reasoning of some imaginary peers (see Evaluate reasoning or solution, Make judgement based on data and Problem based on a real data type of problems in section 2 and [8]). We made the activity interesting for the students by putting it into an authentic context of measuring time in ancient Egypt. The complete problem is shown below.

Type of problem: Evaluate reasoning or solution
Olivia and Jack read that in ancient Egypt people used a water clock or clepsydra to keep time at night. The device consisted of a vessel filled with water, which was then allowed to drain through a small hole in the bottom. Marks on the inside of the vessel marked the precise hours as the water level descends at a constant speed. In the literature, they found the information that one possible shape for the clepsydra was such that the radius r at a particular height h fulfills the equation Olivia and Jack agree that the difference between the measured and calculated time to drain the vessel cannot be attributed to the uncertainties of the measured data. They check in the literature what assumptions have been made in the derivation of the equation and find out that the derivation is based on Bernoulli's equation and the following two assumptions: A1. The velocity of water at the surface is zero. A2. The velocity of the draining water is constant over the whole area of the hole.
All things considered, they explain the observed difference between the measured and calculated time to drain as follows: Jack: »The measured time is longer than calculated because the first assumption is not true. The velocity of the water at the surface is not zero since the water level is moving down.« Olivia: »I think that the observed difference is due to the invalidity of the second assumption. The velocity of the water flowing through the hole is not constant-the velocity is smaller closer to the edge of the hole. The problem is divided into steps that address different learning goals and student difficulties. Steps 1-3 require the students to perform the unit analysis and to explore relations between the independent and dependent variables. In step 4 we present data for the parameters of the experimental setup and the measured time to drain the specific tank. The students have to evaluate the uncertainties of the measured and predicted times to make a judgment about their consistency. In the final step, the students have to evaluate the reasoning of two peers about what assumptions they think are responsible for the discrepancy between the measured data and predicted results. With a hint, we guide the students to check how the invalidity of a given assumption would affect the time. The students need to realize that the time to drain is inversely proportional to the velocity of the draining water and that accounting for the moving water surface yields an increase in velocity of the draining water and therefore cannot be the reason for the observed discrepancy. On the other hand, a non-uniform velocity profile with velocity decreasing towards the edge of the hole results in a reduced flow rate and therefore can be the reason for the discrepancy. The derivation of the given and modified expressions for the time to drain is not required. The emphasis is on developing higher-level reasoning skills. However, the students in calculus-based physics courses might be intrigued to perform the derivation themselves. We asked four master students of our Physics Education Program to solve the problem and to give us detailed feedback about their solving strategies and procedures. The students were solving the problem individually at home and sent us photos of their work. All students performed correct unit analysis, though at first some of them determined the units of a constant c from the expression for the time T (see steps 1 and 2 of the Problem 'Egyptian water clock'). In step 3, the students recognized that the evaluation of the expression for time in step 2 helps them find the ways to change the clepsydra characteristics, or as one of the students wrote 'At step 3 I somehow inferred from the previous findings at step 2 or those findings helped me a lot to solve step 3 faster.'. Some students found this step of the problem to be undefined and would appreciate some guidance for at least the number of suggestions they were supposed to give. One student commented 'I personally think that the instructions for the 3rd step could be more precise (suggest one/two K modifications) because I think that otherwise this step is 'up in the air'. K'. But some others appreciated that the problem was open and gave free rein to their imagination. In step 4 all students concluded that the measured and predicted times were not consistent, though some of the students struggled with the correct determination of the uncertainties of measured and/or predicted time. In the final step, all students realized that the time to drain is inversely proportional to the velocity of the draining water and evaluated accordingly the reasoning of the two peers. When analyzing how the invalidity of a given assumption affects the draining velocity most of the students accounted correctly for the moving water surface and also for the non-uniform velocity profile. Some students performed the derivation of the given expression for the time to drain. In a reflection, the students expressed positive opinions about the problem as one can see from the following comment: 'I found the task very interesting, in particular, step 3 which is very open-ended: first, it offers to modify the clock according to your choice (increase the time/ decrease the time), but there are also possibilities of modifications that are not explicitly visible from the expression for T . This step of the task gives free rein to imagination and allows students with different abilities to come up with a solution, some to more in-depth and some to less-I definitely think that it offers some 'satisfaction', at least I experienced it while solving the task.'. They appreciated that each task yields a hint for the preceding or following task, so if needed they were able to correct their solutions based on new ideas. As one of the students wrote: 'During the activity, with each subsequent step I got a hint to correct the previous one.' As expected, the students found the final step of the problem as the most challenging, however, not overwhelming. Or as students described it: 'I spent most of the time on the last step. When I saw that I was supposed to use Bernoulli's equation, I looked it up to recall what it is about.' or 'I'm pretty sure about the correctness of steps 1 to 4. These steps of the problem were easier than the last one, which was the most time-consuming.'). They also made some useful comments that helped us improve the clarity of text.
We gave the problem also to the same faculty colleagues, who we introduced in section 3, and asked them to solve the problem and to answer the corresponding questionnaire. In the obtained feedback we found patterns similar to the patterns for the electrostatics problem. In the open reflections, three faculty colleagues emphasized the ability to solve problems qualitatively, again one of them associated this ability with intuition. They listed several skills that this problem helps develop, described as 'Testing input parameters, the analysis of limiting cases, dimensional analysis.' or 'The ability of qualitative analysis: estimating values, the effect of parameters, the evaluation of the result.', and also some physics concepts, such as the effect of viscosity on the fluid flow. On the list that we provided, all faculty identified the following two abilities that students can develop by solving this problem and that they would not develop by solving traditional problems: ability to reason/calculate using uncertainties and the ability to critically compare two different explanations. The faculty found this problem useful and instructive, or stated with their own words, 'K The problem is also an exercise in finding alternative hypotheses which is essential for developing scientific approaches.' or 'The problem promotes qualitative thinking about the effects of parameters on the phenomenon-in this way it surpasses most the traditional problems from this topic.'

5.
What could students learn from new types of problems that they do not learn from traditional problems?
First, we realize that it is impossible to give any general answer to this question. There are many factors that determine what students could learn from a problem and there are no sharp lines between traditional and new types of problems. However, we could point to some advantages of the new type of problems by comparing a traditional and a new type of problem, which are designed for the same study level and to solve which the students need to use similar knowledge.
Let us do this by comparing our first problem (force between two charged metal spheres) and the following traditional problem [18]: We wish to determine the electric field at a point near a positively charged metal sphere (a good conductor). We do so by bringing a small test charge, q , 0 to this point and measure the force F 0 on it. Will F q 0 0 / be greater than, less than, or equal to, the electric field E as it was at that point before the test charge was present? What about E at that point when q 0 is there?
The traditional problem above requires from students to make a prediction about the outcome of a hypothetical experiment, and based on this prediction, to compare the electric fields in two scenarios. First, the students need to recall that the test charge is defined as a positive charge. Next, the students need to recall that in metals, negative charges (electrons) are mobile and positive ions are fixed and then to visualize and to predict what happens to the electrons in the metal sphere, when the test charge is near the sphere. The only prompt to make the students think about the influence of the test charge on the distribution of the electrons on the sphere is the word 'near'. There are no dimensions mentioned in the problem that could help the students decide whether the test charge is near enough for the influence to be significant. Based on the predicted change in the charge distribution and using Coulomb's law qualitatively the students could compare forces exerted on the test charge in two scenarios and consequently the electric fields.
The new problem requires the students to identify patterns in data, devise explanations for those patterns and then to use these explanations to match different graphs with the corresponding experiments. In order to identify the patterns, the students first need to read and interpret the graphs. The patterns in the graphs prompt the students to think about the explanations that involve changes in charge distributions. Doing so, students need to recall that in the metals, the electrons are mobile and the positive ions are fixed. When explaining their reasoning, the students are encouraged to draw charge diagrams. In order to decide which graph corresponds to which experiment, the students need to compare the size of the spheres to the distances between them and to apply Coulomb's law in a qualitative form using the charge distributions that they proposed earlier.
We see that while both problems require students to use similar normative knowledge, the new problem engages students in several processes and steps (note the order of the steps) that resemble those commonly used by physicists [19]. By engaging students in such processes, we help students develop their epistemic knowledge, which is an important goal of science education in the 21st century, as we have already discussed in the Introduction.
6. Advice to those who want to design problems using peer-reviewed papers If you decide to design your own problems, using the approach described above, here are some basic advice that might help you: • After you have designed a new problem, write a complete solution, including all answers that require written statements. Try to write the solution the way that you expect your students should do in an ideal case. While writing the solution you may find that some parts are too difficult or too complex. You may decide to add hints or to break those parts into smaller steps or even abandon the idea and start a new one. It may happen that some parts that you first saw as 'trivial' were not that easy and you would realize that you could create a completely new problem only using that part. • Once you have a problem with the solutions, give both to your colleague teachers and ask them for feedback. Ideally, ask a colleague who is familiar with the new type of problems. Colleagues who are not familiar with new types of problems could give valuable feedback regarding physics but their opinion about how difficult or easy the problem would be for students might be incorrect. • Before giving the new problems to your students, you have to model solving a similar type of problems with the students. Modeling solving means that you solve the problem in front of the class, clearly articulating every step and at the end ask students to reflect on the process. Try to replicate the reasoning that you went through when you were first solving the problem, without polishing your initial solution. Let the students see that the reasoning path to the solution of a new problem (even if done by the expert) is not straightforward. • After you give your new problem to the students, collect their work and carefully analyze it (if you are teaching large enrollment classes you might want to give the problem first to a randomly selected small group of students). In addition to this, ask a small number of students (say two best achievers, two middle and two lower achievers) to meet with you and give you feedback on the problem: which steps they found difficult/easy and why, what they think they have learned from the problem, how they liked the problem compared to traditional problem. Student responses are the most valuable piece of information that will show you to which extent the goals of the problem are achieved and will help you improve the problem. Sometimes student responses might reveal an important aspect of the problem that you have not noticed before. For example, when we designed the problem with two charged metal spheres (see section 3) we unintentionally chose positive charges for the case of two equally charged spheres. It was only through observing the groups of students trying to explain the results using only electrons as mobile charges, that we realized how this choice was more challenging than the case of two negatively charged spheres and how it allowed for more opportunities to apply developed knowledge and to deepen it.

Summary
We introduced the reader to the new types of problems that were described before [9] and are becoming an indispensable part of physics education at all levels. We presented two problems of new types that we designed using two peer-reviewed research papers. Both problems are suitable for students at an undergraduate level. The first problem is from electrostatics and the second one is from fluid dynamics. We described the evolution of the problems, student responses to both problems, feedback from our faculty colleagues, and discussed some benefits of new types of problems. Finally, we shared our experience with designing similar problems and provided basic advice to those who wish to design problems using peerreviewed papers. We hope that our paper will encourage physics teachers at universities to start using and designing new types of problems. To inspire those interested, we offer five additional papers [20][21][22][23][24] that in our opinion have the potential to be used to design new problems, as described above.