Chapter 4 D-KEFS Verbal Fluency Test ' ‘ 7 : Overview The D-KEFS Verbal Fluency Test comprises three testing conditions: Letter Fluency Category Fluency, and Category Switching. For the Letter Fluency condition, the exami- i nee is asked to generate words that begin with a particular letter as quickly as possible. For the Category Fluency condition, the examinee is asked to generate words that belong to a designated semantic category as quickly as possible. The last condition, Category ‘Switching, requires the examinee to generate words, alternating between two different semantic categories as quickly as possible. For each trial of each condition, the examinee is allowed 60 seconds. This test measures the examinees ability to generate words flu- ently in an effortful, phonemic format (Letter Fluency), from overleamed concepts (Category Fluency), and while simultaneously shifting between overlearned concepts (Category Switching). The D-KEFS Verbal Fluency Test has a Standard Form and an Alternate Form, In the Standard Form, the Letter Fluency items are F, A, S; the Category Fluency items are Animals and Boys’ Names; and the Category Switching item is Fruits and Furniture. In the Alternate Form, the Letter Fluency items are B, H, R; the Category Fluency items are Clothing and Girls’ Names; and the Category Switching item is Vegetables and Musical Instruments Administration Ages 8-89 a . 2 Materials a Record Form Stimulus Booklet (Easel Position) x Stopwatch Discontinue Do not discontinue the test. Administer all trials of all three conditions in order to derive all of the primary and optional measures. For each trial, discontinue after 60 seconds. 4D-KEFS Verbal Fluency Test Instructions Place and leave the stimulus booklet in its easel position for the entire test. Condition 1: Letter Fluency ise the following prompts during the Letter Fluency condition # If the examinee fails to make a response afier any 1 going. Provide this prompt only once per trial second interval, say, Keep © The first time an examinee generates three consecutive words that do not start with the designated letter, say, The letter we are using now is _. Provide this prompt only once per tial. Keep the stopwatch running while providing prompts, Say, I'm going to say a letter of the alphabet. When | say begin, | want you to tell me as many words as you can that begin with that letter. You will have 60 seconds before | tell you to stop. None of the words can be names of people, or places, ‘or numbers. For example, if | gave you the letter T, you could say take, toy, tooth, and so forth, but you should not say Tom because that is a person's name, you should not say Texas because that is the name of a place, and you should not say twelve because that is a number. Also, do not give me the same word with different endings. For example, if you say take, you should not also say takes and taking. Do you have any questions? Tum the page in the stimulus booklet to display the summarized instructions to the examinee. Say. Here is a page that will help you remember the rules. Review the instructions with the examinee, pointing to the four rules in tun, Leave the sum- marized instructions in the examinee’s view throughout the administration of the Letter Fluency condition. Note: If you are administering the Alternate Form of this test, substitute the letter that appears in brackets in the following instructions. Trial 1 Say, The first letter is F [6]. Ready? Begi Start timing. On the record form, write the examinees responses verbatim in the column labeled “F” (“B” if using the Alternate Form). Record responses that the examinee gener- ates during the first 15 seconds in the first box (labeled “I-15 Seconds”), record responses given in the second 15 seconds in the second box (labeled “16-30 Seconds”) and so forth. After 60 seconds, say, Stop. € é mam amAdministration Trial 2 Introduce the letter by saying, The next letter is A [H]. Ready? Begin. Start timing, Record the examinee’s responses as described for Trial 1, After 60 seconds, say, Stop. Trial 3 Introduce the letter by saying, The next letter is S [A]. Ready? Begin. Start timing. Record the examinee’s responses as described for Trial 1. After 60 seconds, say. Stop. Condition 2: Category Fluency Use the following prompts during the Category Fluency condition: © If the examinee fails to make a response after any 15-second interval, sa going. Provide this prompt only once per trial Keep The first time an examinee generates three consecutive words that do not fit the cate~ gory, say, The category we are using now is __. Provide this prompt only once per trial. Keep the stopwatch running while providing prompts. Note: If you are administering the Alternate Form of this test, substitute the category that appears in brackets in the following instructions. Trial 1 Say, Now we are going to do something a little different. This time, | want you to tell me as many animals [items of clothing] as you can. It doesn’t matter what letter they start with. You will have 60 seconds before | tell you to stop. Do you have any questions? Ready? Be Start timing. On the record form, write the examinee’s responses verbatim in the column. labeled “Animals” (“Clothing” if administering the Alternate Form). As before, write the examinee’s responses in the appropriate 15-second interval sections. At the end of 60 sec- onds, say, Stop.D-KEFS Verbal Fluency Test Trial 2 Introduce the category by saying, Now tell me as many boys’ names [girls’ names] as you can. You will have 60 seconds before | tell you to stop. Ready? Begin. Start timing. Record the examinee’s responses as described for Trial 1. At the end of 60 seconds, say, Stop. Condition 3: Category Switching Use the following prompts during the Category Switching condition © If the examinee fails to make a response after any 15-second interval, say, Keep going. Provide this prompt only once for the Switching condition. © The first time an examinee generates three consecutive words that are not members of one of the two designated categories, say, The categories you are to switch between are__and . Provide this prompt only once for the Switching condition. Keep the stopwatch running while providing prompts. Note: If you are administering the Alternate Form of this test, substitute the categories ‘that appear in brackets in the following instructions (say “musical instruments,” not “pieces of musical instruments”). Say, Now we are going to do something a little different. | want you to switch back and forth between saying as many fruits [vegetables] and as many pieces of furniture (musical instruments] as you can. It doesn’t matter what letter they start with. You will have 60 seconds before | tell you to stop. So you would say a fruit [vegetable], then a piece of furniture [musical instrument], then a fruit [vegetable], then a piece of furniture [musical instrument}, and so on. You can start with either a fruit [vegetable] or a piece of furniture [musical instrument] Do you have any questions? Ready? Begin. Start timing, As before, record the examinee's responses in the appropriate 15-second interval sections. At the end of 60 seconds, sa Stop. Recording Condition 1: Letter Fluency For each trial, record all responses verbatim for each 15-second interval, To obtain scores for each letter trial, sum the following variables across all four 15-second intervals: correct responses, set-loss errors, and repetition errors. To obtain the total raw score for Letter ="’ Administration 59 Fluency, sum the total correct responses for each trial. To obtain the optional interval totals, sum the total correct responses for each 15-second interval across trials and enter the total in the appropriate box at the far right (e.g., the box labeled “I"-15" F +A +S. Correct Responses”). Also enter the total number of correct and incorrect responses gener ated on all three letter trials. Figure 4.1 illustrates these steps D-KEFS Verbal Fluency Test ~ Condition 1: Letter Fluency FE A s tat | amaigcam sable fav ity anagram sat famous austere | saturate em | SatvOUS | | @es2B) slick Intrva: poet — satire "3 | z filthy agrarian | | sanetimoricos: fever aggrandize | | seismograph ferd @array size 2s socons | SUTY antagonize ‘sardy om, wont | festive we] | aggravate 82%! | sondtu = " [ [ seen Co) ack & Ea) friend | gedowes) | sundown | fantastic oseidous separate tosten | cccumiate ae ike | | accumulative Hl ea future additive, }| silty fod gram) -R | sarcastic fastest artdore | sven |e fam | we | fae eae oo ee ‘oan | | [ oe ral B Au wt) eal LIF m [17 eautttenn [0 sauston [3 ] satstttros [_O peowworeros | rwontoneros | t_ | raceme ena [| © | Pa some Se «|| Figure 4.1, Example of Recording Scores for the Letter Fluency Condition of the D-KEFS Verbal Fluency Test for a 72-Year-Old Examinee60 _D-KEFS Verbal Fluency Test Condition 2: Catego For each tial, record all responses verbatim for each 15-second interval. To obtain scores for each category trial, sum the following variables across all four 15-second intervals Correct responses, set-loss errors, and repetition errors. To obtain the total raw score for Category Fluency, sum the total cortect responses for eactt «ial. To abtain the optional interval totals, sum the total correct responses for each 15-second interval across trials and enter the total in the appropriate box at the far right. Also enter the total aumber of correct and incorrect responses generated on both trials. Figure 4.2 illustrates these steps. D-KEFS Verbal Fluency Test (continued) Conon 2: nego Faoey Animats: Boys’ Names ee 1S | ‘se 2 Se = = | [Readies on cory Cospen =. cae, ES fe] win [2] oon LO] Se SSS) Figure 4.2. Example of Recording Scores for the Category Fluency Condition of the D-KEFS Verbal Fluency Test for a 72-Year-Old ExamineeAdministration 61 Condition Record all responses verbatim for each 15-second interval. To obtain total scores for the trial, sum the following variables across all four 15-second intervals: the number of accu- rate switches (record in the Total Switching Accuracy box), the number of correct responses for each category independent of switching accuracy, set-loss errors, repetition errors, and the number of correct and incorrect responses. To obtain the total raw score for Category Switching, sum the total number of correct responses for the two c: To obtain the optional interval totals, enter the total correct responses for each 15-second interval in the appropriate boxes at the far right. Figure 4,3 illustrates these steps. Category Switching tegories. D-KEFS Verbal Fluency Test (continued) Condition 9 Category Switehing Fruits / Furniture | apple 7 ottoman | pum 7 char | xs oronge / soto ms pireapple — / bed watermelon / toctstoo! ding room apie | sume 7 mororeyele tat G (2) + (2) + fe) see —— 3B] stm tree ‘wpm Sere [IE | se Reno Figure 4.3. Example of Recording Scores for the Switching Condition of the D-KEFS Verbal Fluency Test for a 72-Year-Old Examinee6: D-KEFS Verbal Fluency Test Summary of Scores For the primary and optional measures of the D-KEFS Verbal Fluency Test. use the con- version tables in Appendix B. For the primary measures of each condition, enter the total correct responses; for total switching accuracy, enter the number of correct switches, Convert these raw scores to scaled scores. For the contrast measures, enter the scaled scores for each condition in the appropriate boxes and subtract. Convert these scaled-score differences to new contrast scaled scores. For the optional measures, enter the totals for each 15-second interval of each condition and sum these to obtain the total raw score for that interval, For the error measures, sepa rately sum the set-loss errors across all three conditions. To obtain total repetition errors, sum the total repetitions across all three conditions. To obtain the percent set-loss errors raw score, divide the total number of set-loss errors across all three conditions (raw score) by the total number of correct and incorrect responses across all three conditions. Multiply that result by 100. To derive the percent repetition errors raw score, divide the total num- ber of repetition errors across all three conditions by the total number of correct and incor- rect responses across all three conditions. Multiply that result by 100. For Condition 3 only, compute the percent switching accuracy score by adding 1 to the total Switching accuracy raw score and dividing this sum by the total number of correct and incorrect responses for Condition 3 only. Multiply this result by 100. All of the raw scores for these optional measures are converted to scaled scores. Figure 4.4 illustrates these steps. Scoring Condition 1: Letter Fluency The primary measure derived for the Letter Fluency condition is the number of correct words generated within each 60-second trial. A correct response is a word that (a) meets the criteria of the condition (specifically, starts with the designated letter and is not the name of a person, place, or number) and (b) is not a repetition within that trial. The num- ber of correct responses is scored for each of the four 15-second intervals and for the entire 60 seconds of a tial Although most responses generated by examinees are easy to classify as correct or incor- rect, some responses present special, ambiguous scoring problems (e.g., the response sandy generated during the S trial is correct if used as an adjective but incorrect if used as 4 person's name). For this reason, specific guidelines were established for scoring the most common ambiguous responses given by children and adults in the D-KEFS norma- tive sample. The general approach for developing these guidelines was to give the benefit of the doubt to the examinee and to reward “close-call” responses as correct. This general approach reflects the intention to reward evidence of higher-level cognitive skills rather than penalizing examinees for a lack of precision in expressing those skills.—— | _— a Scoring D-KEFS Verbal Fluency Test: Summary of Scores Primary Measures Condon Condition 2 Condon 2 Conaon 3: Later Flency Cateery Perey category Swishing _ Category Swehing Metal Comet “ta Comet Tela Govoct Responses Toll ching Aesurey ellis | oli] i 4 the Primary Contrast Measures Leter Fancy ve. Cetegory Fioney” ay CHAO Sig. Cpe) Fn” Soe Lae tumey: cmgyrmeyseutacre cums ces Cay my tenn cna @-e- wwe afeesfiel= =e] — al * Alo righ onset ays vt opt rtre eaars a Optional Measures: Conditions 1-3 Combined Cconation :Conaion 2: Conaition 3: Letior etegory ‘etegory Fisncy Fluency switching Toul Raw Score «Raw Score «Raw Score Raw Score Scaled Score Fist interval (t"-tsryTotalcorer «(is ] + [ie] + [o] = [as io Second interval (16-30"; Tow Conect [15] + [2] + [4] = (a 16 Thirdinterat at-A5":Towtcorect [SF] + [2] + Li] = [es iz Fourth interval (46'-60"):Total Correct [9 + 7) + (2) = Pe Set-Loss Errors 3 se [ete] Pes cca c Repettion Erors Oo oie eee [eo “otal Responses aan) iy {Coneetessrect eee ae ee Percent Repetition Erors al 3] + [ilt}x we 27) — [i Category Switching: So nice Seow Poreent Switching Accuracy \ecal —~ taal (Gonaiton'3 Ong). Ce) + xw = [666] 1 Figure 4.4, Example of Computing Summary Scores for the D-KEFS Verbal Fluency Test for 4 72-Year-Old Examine64 D-KEFS Verbal Fluency Test Guidelines for Scoring Correct Responses Correct responses are words that (a) start with the target letter; (b) are not names of people or places or numbers; (¢) are not grammatical variants of a word previously given on that trial (e.g., fast, faster); and (d) are not repetitions within that trial In most cases, if a word can be either a common noun or the name of a person or place (e.g. sandy, frank), then the response is considered a common word and scored as correct. On the other hand, if the ambiguous word is given along with other words that are clearly the names of people or places. the response is scored as incorrect. For example, for the S trial, if an examinee says, “Sally, Sandy, Sarah,” the word sandy in this context would be considered the name of a person and scored as an error (see the discussion on incorrect responses). If @ word can be either a common word or a number (¢.g., for), itis scored as a com- ‘mon word unless (a) it is given along with other numbers (¢.g.. “four, five”), or (b) the examinee specifically clarifies that the response is a number. Contractions (e.g., aren't, haven't, hasn't) are counted as correct if the first letter is correct for a particular trial. If an examinee says both the root word of a contraction and the contraction (e.g., “are” and then “aren’t” or “aren't” and then “are”), both are scored as correct. Compound words (¢.g.. air mail) are scored as correct but only as one response. If a response includes multiple compound words in which one of the words is the same (e.g., “apple, applesauce, apple juice, apple pie”), all are counted as correct. Slang words (e.g.. ain't) and swear words that start with the correct letter are scored as correct responses, However, expletives should be noted clinically. Some proper nouns are not names of people or places (e.g., months or days of the week), These proper nouns are counted as correct because the test instructions specifically exclude only those proper nouns that are names of people or places. The names of some months may be confused with names of people (e.g. April). If such a word is given along with the names of other months (e.g., “April, August”), it is scored as correct. On the other hand, if such a word is said along with other names of people (e.g., “April, Annie”), itis scored as incorrect. If such a word is given in isolation (.e., no other months or names of people are given), the response is scored as correct,Scoring Guidelines for Scoring Incorrect Responses Incorrect responses are categorized into two general types: set-loss and repetition errors. Set-Loss Errors A set-loss error is a response that violates any of the criterion rules of the condition. © Words that start with a letter other than the letter designated for the trial (e.g., apple in the F trial) are incorrect responses. If a word starts with a wrong letter but has the same target sound (e.g., phone in the F condition), itis also scored as a set-loss error; however, such responses should be noted clinically Words that start with the target letter but are unambiguous names of people or places or numbers are scored as set-loss errors. Grammatical variants of a word (e.g., fast, faster, fastest) are counted as s2t-loss errors if they do not substantially change the meaning of the root word. Thus, in the example, fast is scored as correct, but faster and fastest are scored as set-loss errors. Grammatical variants of a word that are formed by changing the tense of a verb (e.g. {fool, fooled; sing, sang) or the number of a noun (apple, apples) are scored as set- loss errors. However, if the addition or deletion of a suffix substantially changes the meaning of the root word in some way (e.g.. sing, singer ot fool, foolish), then these pairs of responses should be considered as semantic variants (not grammatical vari- ants) and both scored as correct. Unambiguous number responses are scored as set-loss errors regardless of whether they are cardinal numbers (e.g., jive) or ordinal numbers (e.g...fft). Words like pri- ‘mary ot secondary are not considered numbers because they also have nonnumerical meanings and, thus, are considered correct. @ Non-words that start with the target letter are scored as set-loss errors. If it is unclear whether a response that starts with the target letter is a word or non-word, score it as a correct response. Repetition Errors A repetition error is any response that is repeated within the 60 seconds of a trial. © If.an examinee repeats a correct response, the first occurrence counts as a correct, response and the second occurrence is scored only as a repetition error. [fan exami- nee repeats a word that is a set-loss error, the repeated response is scored as both a set-loss error and a repetition error. Repetitions of a response must be identical to the initial response. If an examinee gives a correct response and then a grammatical variant of that response (¢.2., “snake, snakes” in the S tial), the second word is mot scored as a repetition error because it is a grammatical variant of a previously given word and, thus, is a set-loss error. In other words, repetitions must be exactly the same as the initial response, even in terms of grammatical properties (e.g., plurality).66 D-KEFS Verbal Fluency Test Ifa repeated word has at least two meanings (e.g.. fall, fall) or two words are pro- nounced the same (e.g., See, sea), the second response is scored as a repetition error unless (a) the examinee clarifies one of the same-sounding words (e.g., says, “as in season” for the second fall), (b) the examinee spells one of the words (e.g. or (c) the context of the response clearly reflects a different word (¢.2., “see...sail- boat, saltwater, sea”). The rationale for this guideline is that almost all words have ‘two or more meanings, and, thus, repetition errors can rarely be scored without this guideline, The examinee must clarify only one of the two repeated words for both to be counted as correct (the assumption here is that the examinee would not have clari- fied one word unless he or she was aware that it was different from an earlier word). Condition 2: Category Fluency The primary measure derived for the Category Fluency condition is the number of correct words generated within each 60-second trial. A correct response is a word that (a) meets the criteria of the condition (specifically, is a member of the target category) and (b) is not fa repetition within that trial. The number of correct responses is scored for each of the four 15-second intervals and for the entire 60 seconds of a trial. General Guidelines for Scoring Correct Responses Correct responses in the Category Fluency condition are words that (a) are members of the target category for the trial, (b) are not grammatical variants of a word previ- ously given on that trial (e.g., cat, cats), and (c) are not repetitions of words within that trial. Synonyms (e.g., dog, canine or jacket, coat), are all scored as correct. For a word that has multiple meanings and one of its meanings fits the target catego- 1y, it is scored as correct (¢.g., bat is a correct response for the Animals category). This guideline applies even if the relevant meaning of a word is of lower frequency than other meanings of the same word (c.g., bat used as an animal has a lower fre~ quency than bat used as baseball equipment but is still counted as correct for the Animals category). # A response is correct if it is a specific member of the target category or if it is a sub- ordinate category that can be used to denote a specific member (e.g., fish for the Animals category). Words that represent the same ordinate level as the target category (e.g., animal fen for the Animals trial) are incorrect. Words that are superordinate to the target category also are incorrect. For instance, living things is superordinate to Animals because it also encompasses nonanimal living things (ie., plants), and thus living things is incorrect. Adjectives given either alone or with the name of the target category are incorrect (e.g., carnivorous or carnivorous animal would both be incorrect). Adjectives used to denote different attributes of the same category member are incorrect (e-g., for the Vegetables trial, if the examinee say’, “onion, big onion, little onion,” the first+ Scoring response is correct, but the second and third responses are incorrect because they simply refer to different attributes of the same member). However, if the addition of an adjective results in a new member of the category (e.g., onion, green onion, yel- Jow onion), then all of those words are scored as correct. Guidelines for Animals (Standard Form) ‘Two or more names of different types or breeds of an animal (“collie, rottweiler, boxer” or “goldfish, trout, salmon”) are all scored as correct. Names that represent the genders of the same type of animal (e.g., cow, bull or hen, rooster) are both counted as correct. Words that refer to the same animal at different developmental stages (kitten, cat or foal, colt, horse) are all counted as correct. Specific Guidelines for Clothing (Alternate Form) . « Items typically sold in a clothing store or the clothing department of a department store are considered clothing (e.g., scarf, gloves). Jewelry items (e.g., earrings, necklace), which are typically sold in jewelry stores, are not considered clothing and are scored as incorrect (i.¢., as set-loss errors). Cloth (e.g., wool, fleece, cotton) typically is not sold in bulk form in clothing stores and, thus, when given in isolation are counted as incorrect (i.e., as set-loss errors). The same type of clothing item made of different material (e.g., wool sweater, cotton sweater) refer to different members of the category and are all scored as correct, Different but related members of an article of clothing (e.g., dress shirt, undershirt, T-shirt) are all counted as correct. Specific Guidelines for Boys’ Names (Standard Form) and Girls’ Names (Alternate Form) ° ° Many names are used for both boys and girls (e.g., Chris, Pat, Sandy), These types of names are correct for either category Many names are unusual (c.g., Keilana, Dawayin) and are vague in terms of gender. ‘The examinee is given the benefit of the doubt, and such names are counted as correct. Variations of the same or similar name (e.g. Catherine, Kate, Kathy or William, Bill, Billy) are all counted as correct. Nicknames (e.g., BJ, Junior) are all counted as correct. If the examiner is unsure that a word is a nickname, the examinee is given the benefit of doubt and the response is scored as correct. Language variations of the same name (John, Juan) are all scored as correct. _ Ast #* ¢68 D-KEFS Verbal Fluency Test Guidelines for Scoring Incorrect Responses Incorrect responses are categorized into two general types: set-loss and repetition errors, Set-Loss Errors ‘A set-loss error is a response that violates any criterion rule of the condition. © Words that do not belong to the target category (e.g., the response “fur” given for the Animals tial) are set-loss errors. Grammatical variants of a word (e.g., car, cats) are counted as set-loss errors if they do not substantially change the meaning of the root word. Thus, in the example, cat is scored as correct, but cats is scored as a set-loss error. © Words that are ordinate (e.g., animal) or superordinate (e.g., living things) to the tar- get category are set-loss errors. Repetition Errors ‘A repetition error is any response that is repeated within the 60 seconds of a trial. @ [fan examinee repeats a correct response, the first occurrence is scored as correct and the second occurrence is scored only as a repetition error. If an examinee repeats a word that is a set-loss error, the repeated response is scored as both a set-loss error and a repetition error. © Repetitions of a response must be identical to the initial response. If an examinee gives a correct response and then a grammatical variant of that response (¢.2.. “dog. dogs” in the Animals trial), the second word is not scored as a repetition error because it is a variant of a previously given word and is a set-loss error. In other words, repetitions must be exactly the same as the initial response. even in terms of grammatical properties (e.g.. plurality). Condition 3: Category Switching General Guidelines for Scoring Correct Responses ‘Two types of correct-response scores are derived for the Category Switching condition: total correct responses score and total switching accuracy score. 4 For the total comrect-responses score, the correct responses from each of the two tar- get categories are summed together, regardiess of whether or not the examinee accu- rately followed the switching rule. For the Standard Form, a scaled score is derived for the combined number of fruits given, independent of switching accuracy, and the number of furniture items given, independent of the switching accuracy. For the ‘Alternate Form, a scaled score is derived for the combined number of vegetables given, independent of switching accuracy. and the number of musical instruments given, independent of the switching accuracyScoring @ For the total switching accuracy score, the correct across-category switches are summed. For the Standard Form, the total switching accuracy raw score is the total number of times the examinee gives a fruir response followed immediately by a fir niture response, ot vice versa, For the Alternate Form, the total switching accuracy raw score is the total number of times the examinee gives a vegerable response fol- lowed immediately by a musical instrument response, or vice versa. For a switch to be scored as correct, the examinee must have given consecutive words, first from one target category, then from the other target category. A switch is considered correct even if a word from a target category is a repetition error or a grammatical variant error (e.g., singular and plural forms). For example, if an examinee says “Apple, bed, orange, bed,” the repeated word, bed, is scored as a repetition error, but the switch from orange to bed is still counted as @ correct across-category switch. If the exami- nee switches from a target-category word to a non-target-category word, or vice versa (e.g. “apple, pills” or “pills, apple.” where pills belongs to a nontarget catego- ry), these switches are not counted as correct. Similarly, consecutive within-category responses (e.g., “apples, oranges”) are not counted as correct switches. Specific Guidelines for Fruits (Standard Form) or Vegetables (Alternate Form) @ Some words may technically belong to one category but are commonly believed to belong to the other category (e.g., tomatoes and avocados are often considered veg- tables but they are technically fruits). These types of words are counted as correct for either category. The rationale for this rule is that many people were taught, for instance, that a tomato is a vegetable, and their category-fluency performance should not be penalized because of this previously learned categorization. Related members of the Fruits or Vegetables category are scored as correct because people commonly use these terms to refer to specific members of these categories, For instance, all of the following responses would be scored as correct: “melon, cantaloupe, watermelon”; “berry, blueberry, strawberry”; “onion, red onion, green onion.” Specific Guidelines for Furniture (Standard Form) ‘Related members of the same piece of furniture (e.g., table, end table, dining table) are all counted as correct. @ Items that are commonly sold in a furniture store (e. ered furniture and are scored as correct. .. lamps, rugs, beds) are consid- © Items that are more commonly sold in a store other than a furniture store (e.g., com- puters, televisions, VCRs, phones, blankets, refrigerators, ovens) are not considered furniture and are scored as incorrect. 6970 D-KEFS Verbal Fluency Test Specific Guidelines for Musical Instruments (Alternate Form) @ Related members on the same musical instrument (violin, fiddle, Stradivarius) are all counted as correct. Musical instruments that can also be toys (kazoo, harmonica) are considered correct. Musical instruments that are from other cultures (¢.g., djembe, bagpipe, sitar) are considered correct. Examples of Scoring Correct Responses and Correct Switches ‘Two types of correct responses are scored for the Category Switching condition: (a) the number of initial correct words from each category, independent of switching accuracy, and (b) the number of accurate switches between categories. The following examples illus- trate how these correct responses are scored. In these examples, the symbol ~ is used to denote accurate switches. Example 1: “Apple, ~ desk, ~ banana” Number Correct Fruits 2 Number Correct Furniture 1 Number Accurate Switches Example 2: “Apple, bananas, ~ desk, sofa” Number Correct Fruits D Number Correct Furniture 2 Number Accurate Switches 1 Example 3: “Apple, ~ desk, sofa, ~ orange” Number Correct Fruits 2 Number Correct Furniture 2 Number Accurate Switches 2 Guidelines for Scoring Incorrect Responses Incorrect responses are categorized into two general types: set-loss and repetition errors. Set-Loss Errors |A set-loss error is a response that violates any criterion rule in terms of individual-word generation. & Words that do nor belong to cither target category (¢.2., coffee given for the target switching categories, Fruits/Furniture) are set-loss errors. © Grammatical variants of a given word (¢.g., apple, apples) ate counted as set-loss errors if they do not substantially change the meaning of the root word.Scoring Responses that are ordinate or superordinate to either target category (e.g., fruit or food, respectively, for the Fruits trial) are set-loss errors, Repetition Errors A repetition error is any response that is repeated within the 60 seconds of a trial. © If.an examinee repeats a correct response, the first occurrence is scored as a correct response and the second occurrence is scored only as a repetition error. If an exami- nee repeats a word that is a set-loss error, the repeated response is scored as both a set-loss error and a repetition error. ‘© Repetitions of a response must be identical to the initial response. Examples of Scoring Correct Responses, Correct Switches, and Errors The following examples illustrate how the two types of correct-response measures and the two types of errors are scored. In these examples, the symbol ~ is used to denote accurate switches. Example 1: “Apple, ~ desk, sofa, ~ apple” Number Correct Fruits i Number Correct Furniture Number Accurate Switches Repetition Errors 1 (the second occurrence of apple) Example 2: “Apple, dog, banana, cat, apple, dog” Number Correct Fruit Number Correct Furniture 0. Number Accurate Switches 0 Set-Loss Errors 4 (dog, cat, apples, dog: apples is a grammatical variant of apple) Repetition Errors 1 (the second occurrence of dog) Normed Variables Primary Measures Total-Correct Scores The primary scoring measure for cach of the three conditions of the D-KEFS Verbal Fluency Test is the number of correct responses generated by the examinee, summed across the different trials within each condition (e.g., the three letter trials of the Letter Fluency condition). For Condition 3: Category Switching, two total-correct scores are derived: (a) the total number of correct responses for both target categories summed 1D-KEFS Verbal Fluency Test together, independent of switching accuracy; and (b) the total number of accurate switches from one target category to the other. The raw scores for these four measures are con- verted to scaled scores corrected for each of the 16 age groups (see Appendix B or J). Table 4.1 summarizes these measures. Table 4.1, Primary (Total-Correct) Measures for the D-KEFS Verbal Fluency Test Variable Raw Score Computation __Norming Method _Corrected For Condition 1: Number Correct Responses Scaled Score Each of 16 Lottor Fluency ‘Summed Across (M=10; SD=3) ‘Age Groups Total Correct Three Letter Trials Condition 2: Number Correct Responses Scaled Score Each of 16 Category Fluency ‘Summed Across (M=10; Si Age Groups Total Correct ‘Two Category Trials Condition 3 Number Correct individual Scaled Score Each of 16 Category Switching Responses Summed Across (M=10; SD Age Groups Total Correct Responses’ ‘Two Category Trials Condition 3 Number Correct Switches Scaled Score Each of 16 Category Switching (M=10; SD=3) ‘Age Groups ‘Total Switching Accuracy’ * This measure reflects the total number of correct responses generated for the two target categories summed together, independent of switching accuracy. " This measure quantifies the number of correct switches from one target category tothe other target category, independent of whether the individual items are accurate, repeated, or variant responses. Contrast Measures ‘Some examinees may exhibit disproportionate impairment in Letter Fluency relative to Category Fluency, or vice versa (see the discussion on clinical interpretation). In addition, some individuals may have greater difficulty on Category Switching than on Category Fluency. For these reasons, two contrast scores that directly quartify these comparisons are computed for the D-KEFS Verbal Fluency Test. The scaled-score differences for these contrast measures are converted to new scaled scores corrected for the total sample (see Appendix B or J). Table 4.2 summarizes these contrast measures, Table 4.2. Contrast Measures for the D-KEFS Verbal Fluency Test Contrast Measure Raw Score Computation _Norming Method Corrected For Letter Fluency Condition 1 Scaled Score Contrast Scaled Score Total Sample vs. Minus = ) Category Fluency Condition 2 Scaled Score Category Switching Condition 3 Scaled Score Contrast Scaled Score Total Sample vs. Minus (M=10; SD=3) Category Fluency Condition 2 Scaled Score’ Scoring Optional Measures } The D-KEFS Verbal Fluency Test is composed of three conditions, with a total of six indi- vidual trials: three letter-fluency trials, two category-fluency trials, and one switching trial. i Several process measures are derived and are summed across the three conditions (six trials) of the test: 1 Total number of correct responses generated during each of the 15-second intervals (first 15 seconds, second 15 seconds, etc.), summed across Conditions 1-3 ) Total number of set-loss errors, summed across Conditions 1-3 Total number of repetition errors, summed across Conditions 1-3 @ Percentage of set-loss errors relative to the total number of responses (correct and incorrect) summed across Conditions 1-3 # Percentage of repetition errors relative to the total number of responses (correct and incorrect) summed across Conditions 1-3 In addition, for Condition 3 only, the percentage of accurate switches relative to the total j number of responses (correct and incorrect) on that condition is computed. ‘The raw scores for seven of the optional measures are converted to scaled scores corrected i for each of the 16 age groups: for total set-loss errors, scores were corrected for the total sample, and for total repetition errors, scores were corrected for 9 modified aze groups j (sce Appendix B). Table 4.3 summarizes these measures. Table 4.3. Optional Process Measures for the D-KEFS Verbal Fluency Test ) Raw Score Norming Corrected Variable Conditions Computation Method For 9 Fret inervat _Conditons ‘Total Number Carrct Responses __Scalea Score Eacnot 16 ToialCorect 1-8 in0-15 Secon inte (W=10;S0=9) Age Groups Second itera Contos Total Number Correct Responses Each of 16 D Total Correct 13 in 16-30 Second interval Age Groups “hid Interval Conctions Total Number Cortect Responses Each of 16 Toa Conect 1-8, 'n 31-45 Second Intral Age Groups esac condiccaTeasjuinbe Gonect aiace acho 18 ToalCorect 1-8, in 48-60 Second Interval Age Groups SetLoss Conditions Toil Number Set-Loss Enors Tal Samole 25 Repetition Conditions Total Number Repetition Errors Each of 9 Erors ts Age Groups YD Percent Condtens _Number Setoss Ero ygq__Sealed Soore Each of 16 SetLoss Enos 1-3 Number Total Responses? (=10;80=3) Age Groups ) Percent ‘Conditions Number Repetiton Errors 49) ‘Scaled Score Each of 16 RepetttonErors 1-3 “NurnberTtt Responses! (ue 10,8 Age Groups Percent Condon _Numoer ComeetSutehes + gq Sealed Soore Each ot 16 y Suing Sony Number Total Responses (= 10; 8 AgeGrope ‘Accuracy ‘on Condition 3° “The total number of responses includes both correct and incorrect responses. Note. Some repetition errors are also coded as Y_seeoss errors: each double-coded error counts as only one response forthe toal responses measure. The total number of responses in Condition 3 (Category Switching) includes (a) correct responses regardless of switch securacy and (b) incorrect responses (do not simply sum set-loss and repetition errors because some errors are double-coded),