Strategies for Improving Test Performance: Introduction
There are many strategies for studying and exam taking, and decisions about which ones to use are partly a function of individual habit and preference. However, basic study rules may be applied to any learning exercise; test-taking strategies depend on the type of examination. For those interested in test-writing strategies, the Case and Swanson reference is strongly recommended (see References).
Five Basic Study Rules
1. When studying dense textual material, stop after a few pages to write out the gist of it from memory. If necessary, refer to the material just read. After finishing a chapter, construct your own tables of the major drugs, receptor types, mechanisms, and so on, and fill in as many of the blanks as you can. Refer to tables and figures in the book as needed to complete your notes. Create your own mnemonics if possible. Look up other mnemonics in books if you can't think of one yourself. These are all active learning techniques; mere reading is passive and far less effective unless you happen to have a photographic memory. Your notes should be legible or typed on a computer, and saved for ready access when reviewing for exams.
2. Experiment with other study methods until you find out what works for you. This may involve solo study or group study, flash cards, or text reading. You won't know how effective these techniques are until you have tried them.
3. Don't scorn "cramming," but don't rely on it either. Some steady, day-by-day reading and digestion of conceptual material is usually needed to avoid last-minute indigestion. Similarly, don't substitute memorization of lists (eg, the Key Words list, Appendix I) for more substantive understanding.
4. If you are preparing for a course examination, make every effort to attend all the lectures. The lecturer's view of what is important may be different from that of the author of a course textbook, and chances are good that exam questions will be based on the instructor's own lecture notes.
5. If old test questions are legitimately available (as they are for the USMLE and courses in most medical schools), make use of these guides to study. By definition, they are a strong indicator of what the examination writers have considered core information in the recent past (also see Point 4).
Strategies Applicable to All Examinations
Three general rules apply to all examinations.
1. When starting the examination, scan the entire question set before answering any. If the examination has several parts, allot time to each part in proportion to its length and difficulty. Within each part, answer the easy questions first, placing a mark in the margin by the questions to which you will return. Practice saving enough time for the more difficult questions by scheduling 1 min or less for each question on practice examinations such as those in Examinations 1 and 2 in the Self-Assessment. (The time available in the USMLE examination is approximately 55–60 sec per question.)
2. Students are often advised to avoid changing their first guess on multiple-choice questions. However, research has shown that students who are unsure of the answer to a question make a change from the incorrect answer to the correct answer about 55% of the time. So if you are unsure of your first choice for a particular question and on further reflection see an answer that looks better, research supports your making one—but only one—change.
3. Understand the method for scoring wrong answers. The USMLE does not penalize for wrong answers; it scores you only on the total number of correct answers. Therefore, even if you have no idea as to the correct answer, make a guess anyway; there is no penalty for an incorrect answer. In other words, do not leave any blanks on a USMLE answer sheet or computer screen. Note that this may not be true for some local examinations; some scoring algorithms do penalize for incorrect answers. Make sure you understand the rules for such local examinations.
Strategies for Specific Question Formats
A certain group of students—often characterized as "good test-takers"—may not know every detail about the subject matter being tested but seem to perform extremely well most of the time. The strategy used by these people is not a secret, although few instructors seem to realize how easy it is to break down their questions into much simpler ones. Lists of these strategies are widely available (eg, in the descriptive material distributed by the National Board of Medical Examiners to its candidates). A paraphrased compendium of this advice is presented next.
Strategies for the "Choose the One Best Answer" (of 5 Choices) Type Question
1. Many of the newer "clinical correlation" questions on the Board exam have an extremely long stem that provides a great deal of clinical data. Much of the data presented may be irrelevant. The challenge becomes one of finding out what is being asked. One method for rapidly narrowing the search, especially when confronted with a long stem, is to just read the last sentence of the stem, then scan the answer list. The nature of the last sentence and the answers provide a clue to the parts of the stem that are relevant and those that are not.
2. If 2 statements are contradictory (ie, only 1 can be correct), chances are good that 1 of the 2 is the correct answer (ie, the other 3 choices may be distracters). For example, consider the following:
1. In treating quinidine overdose, the best strategy would be to
(A) Acidify the urine
(B) Administer a calcium chelator such as EDTA
(C) Alkalinize the urine
(D) Give potassium chloride
(E) Give procainamide
The correct answer is A: acidify the urine.
In the pair of contradictory distractors (choices A and C), the instructor revealed what was being tested and then used the other three as "filler." Therefore, if you don't know the answer, you are better off guessing A or C (a 50% success probability) than A or B or C or D or E (a 20% success probability). Note that this strategy is valid only if you must guess; many instructors now introduce contradictory pairs as distracters. Another "rule" that should be used only if you must guess is the "longest choice" rule. When all the answers in a multiple-choice question are relatively long, the correct answer is often the longest one. Note again that sophisticated question writers may introduce especially long incorrect choices to foil this strategy.
3. Statements that contain the words "always," "never," "must," and so on are usually false. For example,
Acetylcholine always increases the heart rate when given intravenously because it lowers blood pressure and evokes a strong baroreceptor-mediated reflex tachycardia.
The statement is false because, although acetylcholine often increases the heart rate, it can also cause bradycardia. (When given as a bolus, it may reach the sinus node in high enough concentration to cause initial bradycardia.) The use of trigger words such as "always" and "must" suggests that the instructor had some exception in mind. However, be aware that there are a few situations in which the statement with a trigger word is correct.
4. Choices that do not fit the stem grammatically are usually wrong. For example:
1. A drug that acts on a 
-receptor and produces a maximal effect that is equal to one half the effect of a large dose of isoproterenol is called a
(A) Agonist
(B) Analog of isoproterenol
(C) Antagonist
(D) Partial agonist
The use of the article a at the end of the stem rather than an implies that the answer must start with a consonant (ie, choice D). Similar use may be made of disagreements in number. Note that careful question writers avoid this problem by placing the articles in the choice list, not in the stem.
5. A statement is not false just because changing a few words will make it somewhat more true than you think it is now. "Choose the one best answer" does not mean "Choose the only correct statement."
Strategies for "All of the Following Are Accurate Except" Questions
1. This type of question is avoided now on the USMLE because of problems with ambiguity; however, this type still is used in many local examinations because question-writers perceive them to be relatively easier to construct. When faced with this type of question, approach it as a nested set of true/false questions in which (hopefully) only one is true. It may help to mark each choice as either "T" or "F" as you read through them.
2. If 2 statements are contradictory, then 1 of them is certain to be the correct (false) answer because they cannot both be accurate statements, and yet this type of question cannot have 2 false answers. For example, consider the following:
1. All of the following may result from the use of thiazide diuretics EXCEPT
(A) Hyperglycemia
(B) Hypernatremia
(C) Hyponatremia
(D) Hyperuricemia
(E) Metabolic alkalosis
The correct answer is B: hypernatremia.
The possibility that thiazide diuretics do not affect serum sodium concentrations is not tenable because that would produce 2 false choices. It is also somewhat unlikely that a drug could cause 2 opposite effects: therefore, the probability that 1 of the opposites is the correct (false) answer is high.
3. If the choices contain 2 drugs that are highly similar, then neither is likely to be the correct (false) answer. For example, consider the following:
1. A young man who had become physiologically dependent after illicit use of secobarbital is undergoing severe withdrawal symptoms, including nausea, vomiting, delirium, and periodic seizures. Which one of the following drugs will NOT alleviate these symptoms?
(A) Buspirone
(B) Chlordiazepoxide
(C) Diazepam
(D) Midazolam
(E) Phenobarbital
The correct answer is A: buspirone.
If you recognize that chlordiazepoxide, diazepam, and midazolam all are benzodiazepine drugs with virtually identical pharmacologic effects, then you can quite safely rule out all three of them.
Strategies for Matching Type Questions
Matching questions usually test name recognition, and the most efficient approach consists of reading each stem item and then scanning the list of choices from the start and picking the first clear "hit." This is especially important on extended matching questions in which just reading the list can be time consuming. (Note, however, that the strategy suggested by the National Board of Medical Examiners for the USMLE differs from the above; see their General Instructions publication.) Occasionally, the strategies described above for the single best answer type question can be applied to the matching and extended matching type.
Strategies for the "Answer A if 1, 2, and 3 Are Correct" Type Question
This type of question, known as the "K type," has been dropped from the USMLE and therefore is no longer represented among the practice questions provided in this Review. However, it is still used in many local examinations.
For this type of question, one rarely must know the truth about all 4 statements to arrive at the correct answer. The instructions are to select
(A) if only (1), (2), and (3) are correct;
(B) if only (1) and (3) are correct;
(C) if only (2) and (4) are correct;
(D) if only (4) is correct;
(E) if all are correct.
Useful strategies include the following:
1. If statement 1 is correct and 2 is wrong, the answer must be B (ie, 1 and 3 are correct). You don't need to know anything about 3 or 4.
2. If statement 1 is wrong, then answers A, B, and E are automatically excluded. Concentrate on statements 2 and 4.
3. The converse of 1 above: If choice 1 is wrong and 2 is correct, the answer must be C (ie, 2 and 4 are correct).
4. If statement 2 is correct and 4 is wrong, the answer is A (ie, 1 and 3 must be correct and you need not even look at them). (See example below.)
5. If statements 1, 2, and 4 are correct, the answer must be E. You need not know anything about 3.
6. Similarly, if statements 2 and 3 are correct and 4 is wrong, the answer must be A, and statement 1 must be correct.
7. If statements 2, 3, and 4 are correct, then the answer must be E, and statement 1 must be correct.
No doubt, more of these rules exist. In general, if you know whether 2 or 3 of the 4 statements in each question are right or wrong (ie, 50–75% the material), you should achieve a perfect score on this kind of question. The best way to learn these rules is to apply them to practice questions until the principles are firmly ingrained.
Consider the following question. Using the above rules, you should be able to answer it correctly even though there is no reason why you should know anything about the information contained in 2 of the 4 statements. The answer follows.
Which of the following statements is (are) correct?
1. The "struck bushel" is equal to 2150.42 cubic inches.
2. Medicine is one of the health sciences.
3. The fresh meat of the Atlantic salmon contains 220 IU of vitamin A per 100 g edible portion.
4. Hippocrates was the founder of modern psychoanalysis.
The answer is A. Because statement 2 is clearly correct and 4 is just as patently incorrect (let's give Freud the credit), the answer can only be A, and statements 1 and 3 must be correct. (The data are from Lentner C, editor: Geigy Scientific Tables, 8th ed. Vol. 1. Ciba-Geigy, 1981.)