Analyzing Error Perception and Recognition Among Professional Communication Practitioners and Academics

Researcher’s note (June 2026). Lindsay Emory Moore and I published this in 2018 because I wanted to know, empirically, whether professional communicators and academics actually agree on what counts as an “error” — so we measured how 303 of them perceived and recognized 24 usage errors. What we found is that botheration is uneven and context-dependent: practitioners were often more bothered than academics, and gender, job type, and years in the field all shifted the picture. That work sits directly underneath my current research on AI in writing and assessment, where the same question now scales up — when a machine flags or fixes “error,” whose judgment is it encoding, and does it match what real readers in a given context actually care about?

Abstract

We investigated the perception and recognition of errors in a population of practitioners and academics in professional and technical communication. Specifically, we measured 303 participants’ botheration levels of 24 usage errors and then correlated those results against their ability to recognize the errors. Results indicated that practitioners were often more bothered by errors than academics and that participants’ overall botheration level might have fluctuated over the past 40 years. Participants’ botheration level also appeared to associate with their ability to identify error. Finally, we found that participants’ gender, job type, and years working in the field influence their error perception.

Keywords: editing tests, error taxonomies, grammar, technical editing, usage

The continued study of usage error reflects language evolution within a community as well as marks our own language biases and character judgments. Previous research in business and professional communication has focused on the error perception and recognition differences between academics and nonacademic professionals (Beason, 2001; Boettger, 2014; Brandenburg, 2015; Gilsdorf & Leonard, 2001; Gray & Heuser, 2003; Hairston, 1981; Leonard & Gilsdorf, 1990).

The resulting research challenged assumptions about classroom grammar teaching: English teachers were prescriptivists who taught language in isolation and only categorized usage as correct or incorrect. This approach did not prepare students for workplace writing where practitioners were less bothered by usage error and more concerned with the overall message. As a result, practitioners’ application of error was often inconsistent or incorrect.

In fact, academics and practitioners are both bothered by error, yet to varying degrees and by different error types (Boettger, 2014; Gilsdorf & Leonard, 2001; Hairston, 1981; Kantz & Yates, 1994; Leonard & Gilsdorf, 1990). Inconsistencies in participants’ error recognition have also been identified, but some of the methodological designs used to collect these data have been challenged and subsequently refined (Brandenburg, 2015; Gray & Heuser, 2003; Williams, 1981). Furthermore, social variables appear to correlate with error perception. For example, females are generally more aware of—and more bothered by—usage errors than males; however, age, experience level, and native language could also influence perception (Gilsdorf & Leonard, 2001; Gray & Heuser, 2003; Hairston, 1981; Kantz & Yates, 1994; Wolfe, Shanmugaraj, & Sipe, 2016).

In this article, we report the results of an error perception and recognition survey. Specifically, we investigated which errors bothered 303 practitioners and academics working in professional and technical communication. In addition, we examined how participants’ botheration level of 24 usage errors correlated with their recognition of these same errors as well as the participants’ gender, job type, and years of experience in professional communication.

We begin by summarizing the results of previous error taxonomies and how different populations and methodological designs have informed our current understanding of error. The “Method” section describes the design and content of our survey. We organize our results around each research question and conclude with directions for future research and implications for the classroom.

Literature Review

Error research in business and professional communication has primarily focused on perception and recognition. Within those studies, the social variables of participants have been shown to influence participants’ botheration of specific error types.

On Error Perception

Most of the relevant research has assessed practitioners’ perceptions of errors (Beason, 2001; Gilsdorf & Leonard, 2001; Gray & Heuser, 2003; Hairston, 1981; Leonard & Gilsdorf, 1990). Collectively, these studies can be classified as the botheration studies, and the results are often consistent; fragments, run-on sentences, misspellings (specifically homophones), subject/verb agreements, dangling modifiers, and unnecessary shifts in verb tense ranked as some of the most bothersome errors.

These findings have offered an excellent resource for classroom instruction, but the methodological designs of the studies separated error from an authentic context. Data were collected via a survey. Researchers placed one usage error within a sentence, and participants registered their botheration on a Likert-type scale. In “The Phenomenology of Error,” Williams (1981) challenged this methodological design, arguing that participants would register a more conservative botheration level if they knew they were being tested and if they were primed to look for one error within each sentence. Even Hairston (1981), who conducted the first botheration study, conceded that her participants were likely self-consciously looking for the error in each sentence and that there was a certain amount of “trying to do well on the test” (p. 798).

Conversely, the errors included in the botheration surveys were often selected arbitrarily and based on the researchers’ perspectives of what could bother professionals. Williams (1981) called for a reliable source of errors that was more context derived. Connors and Lunsford (1988) answered this call with their taxonomy of the 20 most common errors found in college writing. The list was generated from the manual coding of more than 3,000 student essays and has become the primary taxonomy for teaching error in composition classes. The updated Lunsford and Lunsford (2008) taxonomy included new errors that were related to evolving text types and technological advances, illustrating how context influenced the error types that students included in their writing.

Boettger (2011, 2012, 2014) took a similar approach to examining error within professional and technical communication. He derived his taxonomy from a corpus of editing tests used to screen prospective communicators. He hypothesized that the errors hiring managers purposely placed in these tests were also the errors prioritized within professional settings. His taxonomy of 71 usage errors included error types related to content, style, and design that were not in the college writing taxonomies or measured in any of the botheration studies cited earlier. The presence of these additional error types increased the need to investigate how professional communication practitioners perceive error and how those perceptions vary from academics, particularly those teaching professional communication.

On Error Recognition

A smaller number of studies have assessed practitioners’ recognition of error.

Several of the botheration studies addressed the methodological design issues related to error recognition and have generated additional findings as a result (Gilsdorf & Leonard, 2001; Gray & Heuser, 2003; Leonard & Gilsdorf, 1990). For example, Gray and Heuser (2003) included two sentences with the same error to measure professionals’ consistency with recognition. They also included sentences with no errors as well as the category “no error” in their botheration scale. They reported a “disturbing finding” that a number of participants marked “no error” in all but one of the error categories. A percentage of respondents also registered botheration with the grammatically correct sentences. Professionals were also inconsistently bothered by errors; for example, 60% were bothered by one sentence with a subject-verb agreement error, but only 28% of those same participants were bothered by the second sentence (Gray & Heuser, 2003, p. 59). These results begin to problematize if participants can recognize specific errors or if external factors, such as sentence structure or subject matter, influence those perceptions.

Brandenburg (2015) explored if practitioners could recognize errors when they were not intentionally looking for them. She placed six errors in a memo (rather than isolating each error in a sentence) and found that readers only identified one third of the errors. Furthermore, Brandenburg reported difficulties in determining if participants recognized the errors; they tended to describe them rather than identify them by name, complicating the coding process. What Brandenburg did code, though, suggested that respondents were bothered by the same errors that previous studies reported—fragments, word endings, and misspellings related to homophones—which suggests a correlation between error perception and recognition.

On Social Variables

The social variables of participants also appear to influence error perception.

Generally, academics are more bothered by errors than practitioners (Gilsdorf & Leonard, 2001; Hairston, 1981; Leonard & Gilsdorf, 1990). However, these results are typically found in populations of academics in business and professional communication and practitioners from a number of industries. For example, Hairston’s (1981) study included practitioners who represented 64 different occupations.

Females also appear more bothered by error than males; female academics, in particular, are more aware of and more accurate in detecting errors than their male counterparts (Gilsdorf & Leonard, 2001; Gray & Heuser, 2003; Hairston, 1981; Kantz & Yates, 1994). Linguistics research demonstrates that females are more likely to choose more formal registers than males (Finegan & Biber, 2001; Fries, 1940), and the consistency of these gender results across error studies suggests some generalizability.

However, there are other social variables to consider when examining error perception as well as how these variables might interact. For example, Gilsdorf and Leonard (2001) reported that their younger participants were less bothered by usage errors than older participants. Many of these respondents attributed this lower botheration to changes in specific usage elements. Social change impacts language evolution, and it has been hypothesized that people are less bothered by usage error today than they were 40 years ago (Noguchi, 1991). In fact, evidence in error research suggests this change has occurred among nonacademic professionals (Gray & Heuser, 2003; Wolfe et al., 2016).

Our study accounts for these various perspectives. We investigated the perception and recognition of errors in a population of practitioners and academics in professional and technical communication. Specifically, we measured participants’ botheration levels of 24 error types and then correlated those results against participants’ ability to recognize those errors. We also considered how gender, job type, and years working in the field interacted with error perceptions. The study focused on the following research questions:

Research Question 1: Which errors bother professional communication practitioners and academics?

Research Question 2: How do participants’ botheration levels correlate with their ability to identify errors?

Research Question 3: Do social variables influence participants’ botheration levels? Specifically, are participants’ botheration levels influenced by their gender, job type, or years working in the field?

Method

To address our questions, we developed a survey in Qualtrics. The survey asked participants to register their botheration levels of 24 usage errors on a 7-point Likert-type scale. Practitioners were asked to consider each error as if they were evaluating the writing of a prospective employee. Academics were asked to consider each error as if they were evaluating the writing of a senior-level student. The 303 participants were also asked to identify the usage error in each sentence.

Design

Our survey instrument addressed validity and reliability issues raised by Williams (1981) and others about measuring error perception and identification. We selected the 24 errors based on previous research rather than our perceptions of which errors might bother participants. These error types were identified from Boettger’s (2014) editing test taxonomy, which included 71 different errors that were also classified by six broad categories: grammar, punctuation, spelling, content, style, and design. Each category was represented in our survey with four error types. We included the two highest ranked errors, the median error, and the lowest ranked error in each category to provide a more holistic representation of each broad category. The appendix includes the sentences used in the survey.

Each error was represented by one sentence, and only one error was included per sentence. We used or modified sentences from existing textbooks, handbooks, and articles from business, professional, and technical communication (Alred, Brusaw, & Oliu, 2012; Amare, Nowlin, & Weber, 2011; Beason, 2001; Gilsdorf & Leonard, 2001; Gurack & Hocks, 2009; Locker & Kienzler, 2012; Lunsford, Matsuda, & Tardy, 2010; Rew, 1999; Rude & Eaton, 2011; Smith & Bernhardt, 1997). Using existing, published examples acknowledged some validity issues because it assumed these examples best illustrated the error types.

To further enhance the validity of the survey, the 24 questions were randomized; no participants saw the same question in the same order. Participants were also randomly shown one of two possible sentences for each of the 24 usage errors. This approach helped to ensure that participants were responding to the error in the sentence rather than just a poor or confusing sentence.

Participants registered their botheration levels on a 7-point Likert-type scale that ranged from not bothersome (1) to extremely bothersome (7). Participants were also asked to identify each error type and instructed to write “don’t know” if they could not identify the error. We chose a 7-point scale to give participants more latitude to respond in a way that more accurately reflected their perceptions. A scale with fewer points could arbitrarily limit participants and could lead to ceiling or floor effects (e.g., a high number of participants respond either high or low, reducing data variability). Our results cannot be directly compared with the results from previous taxonomies that used 3-, 4-, or 5-point scales; however, they can suggest if error perception has changed.

Participants

Three hundred and three practitioners and academics participated in the survey. Previous botheration studies have typically surveyed academics from business and professional communication but practitioners from a variety of industries. To offer a more comparable sample, we only solicited participants working in or teaching professional or technical communication.

Practitioners and academics were recruited through forums and listservs on professional and technical communication, including the Association for Business Communication forum, Copyediting-L, the Technical Editing Special Interest Group list (through the Society for Technical Communication), the Academic Special Interest Group list (through the Society for Technical Communication), the Association of Teachers of Technical Writing (ATTW-L), and the Council for Programs in Technical and Scientific Communication (CPTSC-L). The study was approved by the researchers’ institutional review board, and participants registered informed consent before beginning the survey. No identifiable information was collected, including IP addresses; however, participants were asked to volunteer demographic information (e.g., gender, years working in the profession). No compensation was provided for participation.

Practitioners accounted for 62.4% of the sample (n = 189), and the academics accounted for 37.6% (n = 114). Most participants were female (60.7%, n = 184). Additionally, 13.9% of participants (n = 42) had either worked in or taught professional communication for 1 year to 5 years, 18.8% (n = 57) for 6 to 10 years, 16.5% (n = 50) for 11 to 15 years, 17.5% (n = 53) for 16 to 20 years, 11.5% (n = 35) for 21 to 25 years, 13.2% (n = 40) for 26 to 30 years, and 8.6% for (n = 26) 31 or more years.

Error Perception Results

Our first research question examined which errors bothered professional communication practitioners and academics. Overall, participants’ botheration levels ranged from somewhat bothersome to moderately bothersome (or 3 to 5 points on the 7-point Likert-type scale). Mann–Whitney U tests identified seven significant differences in error perception between the participant types. Table 1 organizes these results by the three botheration tiers and in descending order of overall mean.

The moderately bothersome tier included nine error types, which represented all six of the broad error categories. Overall, participants were most bothered by sentence fragments (M = 5.69, SD = 1.54; see survey sentences in the appendix). Additionally, academics were significantly more bothered by fragments than practitioners (U = 8414, p = .01). The sentence fragment was the only error that significantly bothered academics more than practitioners.

Table 1. Error Perception Results Organized by the Botheration Tiers and in Descending Order of Overall Mean.

Error type	Broad error	Category, M (SD)	Practitioner, M (SD)	Academic, M (SD)
Moderately bothersome tier
ᵃSentence fragment	Grammar	5.69 (1.54)	5.51 (1.66)	6.03 (1.22)
Unnecessary or missing apostrophe	Punctuation	5.68 (1.49)	5.70 (1.47)	5.64 (1.52)
Misspelling (homophone)	Spelling	5.61 (1.58)	5.73 (1.50)	5.39 (1.70)
Fancy tone/language	Style	5.20 (1.82)	5.28 (1.80)	5.07 (1.86)
Wrong word	Content	5.18 (1.84)	5.13 (1.84)	5.21 (1.84)
ᵃLogic/sequence error	Content	5.13 (1.90)	5.43 (1.70)	4.59 (2.11)
ᵃSpace missing or needed	Design	5.10 (1.71)	5.28 (1.57)	4.77 (1.91)
Incorrect or missing question mark	Punctuation	5.08 (1.86)	5.07 (1.90)	5.09 (1.79)
Poorly integrated source material	Content	5.04 (1.90)	5.07 (1.81)	4.97 (2.08)
Neutral/no opinion tier
Misspelling (general)	Spelling	4.63 (1.90)	4.71 (1.86)	4.49 (1.97)
ᵃIndentation error	Design	4.58 (1.75)	4.78 (1.69)	4.22 (1.82)
Unnecessary shift in verb tense	Grammar	4.58 (1.90)	4.72 (1.82)	4.32 (2.01)
Unnecessary or missing capitalization	Grammar	4.46 (1.98)	4.41 (1.98)	4.57 (1.98)
Vague or missing language	Content	4.44 (2.12)	4.50 (2.15)	4.31 (2.08)
Incorrect alignment	Design	4.25 (2.45)	4.42 (2.50)	3.93 (2.32)
Sentence ends with preposition	Grammar	4.11 (2.23)	4.19 (2.28)	3.97 (2.12)
ᵃMissing comma after an intro element	Punctuation	4.06 (1.81)	4.23 (1.79)	3.75 (1.83)
Incorrect number format	Style	4.01 (1.89)	3.99 (1.84)	4.06 (1.98)
Somewhat bothersome tier
Unnecessary negative construction	Style	3.96 (2.02)	3.94 (1.97)	3.97 (2.12)
ᵃIncorrect text styling	Design	3.94 (1.79)	4.11 (1.79)	3.66 (1.77)
Misspelling (proper noun)	Spelling	3.88 (2.29)	3.96 (2.36)	3.72 (2.16)
Redundant, expendable, or incomparable language	Style	3.81 (1.83)	3.88 (1.82)	3.70 (1.87)
Hyphen, em-dash, or en-dash error	Punctuation	3.54 (1.88)	3.61 (1.87)	3.42 (1.88)
ᵃMisspelling (British)	Spelling	3.44 (1.87)	3.72 (1.82)	2.94 (1.87)

ᵃSignificant at ⩽.05 alpha level.

Practitioners were more bothered by logic or sequence errors than academics (U = 7841, p = .00) as well as missing or needed space (U = 8677, p = .05). These errors were content and design related, respectively, which were identified in Boettger’s (2014) taxonomy and not measured in previous botheration studies.

The neutral/no opinion tier included nine error types, which represented all six of the broad categories. Practitioners were significantly more bothered by indentation errors than academics (U = 7787.5, p = .01) as well as by missing commas in introductory elements (U = 8063.5, p = .05). Indentation error was identified by Boettger (2014) and not measured in previous botheration studies.

Finally, the somewhat bothersome tier contained six error types, which represented four of the six broad error categories. Practitioners were significantly more bothered by incorrect text styling than academics (U = 8404, p = .03) as well as the British spelling of words (U = 6633, p = .00). Incorrect text styling is another design-related error identified by Boettger (2014) and not measured in previous botheration studies.

Error Recognition Results

Our second research question examined how participants’ perceptions correlated with their ability to recognize the 24 usage errors. We yielded results by manually coding the identification responses into three categories: correct identification, correct description of the error, and don’t know/incorrect identification (see Table 2).

On average, participants correctly identified or described the usage errors 60% of the time. Space missing or needed, indentation error, and sentence fragment were correctly identified 51% or more of the time. Similarly, homophone misspelling, incorrect text styling, incorrect alignment, unnecessary or missing apostrophe, wrong word, and logic/sequence error were correctly described by the participants 51% or more of the time.

Table 2. Error Recognition Results Organized by Broad Error Category.

Error type	Broad error	Correct ID, % (N)	Correct, no ID, % (N)	Incorrect, % (N)	Frequency
Sentence ends with preposition	Grammar	45.45 (125)	28.73 (79)	25.82 (71)	275
Sentence fragment	Grammar	51.82 (142)	30.66 (84)	17.52 (48)	274
Unnecessary or missing capitalization	Grammar	43.01 (120)	14.34 (40)	42.65 (119)	279
Unnecessary shift in verb tense	Grammar	11.99 (32)	16.85 (45)	44.57 (190)	267
Hyphen, em-dash, or en-dash error	Punctuation	37.78 (102)	8.15 (22)	54.07 (146)	270
Incorrect or missing question mark	Punctuation	47.81 (131)	32.85 (90)	19.24 (53)	274
Missing comma after an intro element	Punctuation	37.41 (104)	21.22 (59)	41.37 (115)	278
Unnecessary or missing apostrophe	Punctuation	27.80 (77)	57.76 (160)	14.44 (40)	277
Misspelling (British)	Spelling	33.58 (92)	7.30 (20)	59.12 (162)	274
Misspelling (general)	Spelling	24.91 (69)	22.02 (61)	53.07 (147)	277
Misspelling (homophone)	Spelling	14.44 (40)	70.76 (196)	14.80 (41)	277
Misspelling (proper noun)	Spelling	20.44 (56)	12.04 (33)	67.52 (185)	274
Logic/sequence error	Content	8.59 (22)	51.56 (132)	39.84 (102)	256
Poorly integrated source material	Content	7.86 (21)	18.73 (50)	73.41 (196)	267
Vague or missing language	Content	11.83 (33)	30.46 (85)	57.71 (161)	279
Wrong word	Content	14.49 (40)	54.71 (151)	30.80 (85)	276
Fancy tone/language	Style	26.37 (72)	34.07 (93)	39.56 (108)	273
Incorrect number format	Style	23.05 (65)	45.74 (129)	31.21 (88)	282
Redundant, expendable, or incomparable language	Style	12.08 (32)	23.40 (62)	64.53 (171)	265
Unnecessary negative construction	Style	40.00 (110)	16.36 (45)	43.64 (120)	275
Incorrect alignment	Design	20.00 (55)	58.55 (161)	21.45 (59)	275
Incorrect text styling	Design	4.07 (11)	58.89 (159)	37.04 (100)	270
Indentation error	Design	53.62 (148)	26.45 (73)	19.93 (55)	276
Space missing or needed	Design	65.95 (184)	12.54 (35)	21.51 (60)	279

Participants typically could not identify or incorrectly identified the usage errors 40% of the time. Fifty percent or less of the time, participants incorrectly identified poorly integrated source material; unnecessary shift in verb tense; proper noun misspelling; redundant, expendable, or incomparable language; British spelling; vague or missing language; hyphen, em-dash, or en-dash error; and general misspelling.

To further explore these descriptive statistics, we subjected the data to a correspondence analysis (CA). CA is not an inferential measure, and therefore does not determine statistical significance. The statistical output provides a chi-square value, which relates to the overall interaction between the rows and columns; it is up to the researcher to consult other statistical output to properly interpret the results (Boettger & Lam, 2013; Lam, 2014). Results from the CA identified a significant relationship between error type and error identification (χ² = 1883.642, p < .0001), and 10 relationships were identified (see Figure 1).

Results indicated that indentation error, sentence ending with a preposition, sentence fragment, and incorrect or missing question mark corresponded with participants’ correct identification. Homophone misspelling, wrong word, and incorrect alignment corresponded with participants’ correct description. Finally, proper noun misspellings; redundant, expendable, or incomparable language; and vague or missing language were associated with participants’ incorrect identification.

[Figure 1: Correspondence analysis between error type and error identification. Note. The eigenvalues for the first two dimensions are 54.04% and 45.96%, respectively, indicating that this 2-dimensional visualization explains 100% of the variation (inertia). — image to be added.]

Social Variables Results

Our final research question examined if error perception interacted with participants’ gender, job type, and years working in professional communication. For this analysis, we analyzed the average botheration ratings of each of the six broad error categories against participants’ social variables. The methods applied were analysis of variance and Tukey’s test (the latter was estimated to examine individual differences).

Statistically significant results were found among all variables. Overall, female participants were more bothered by grammar (F = 6.968, p = .009), punctuation (F = 7.814, p = .01), spelling (F = 7.473, p = .01), and content (F = 6.661, p = .01) errors than male participants. Gender did not appear to influence participants’ botheration with style or design errors. However, design errors interacted with job type (F = 6.320, p = .01). Practitioners reported a higher botheration with these errors, on average, relative to academics (p = .02). Finally, overall statistical significance was identified with participants’ botheration of grammar error, their gender, and the number of years they have worked in professional and technical communication (F = 2.459, p = .03). On average, male respondents with 30 or more years of work experience reported lower grammar botheration scores than their female colleagues by 0.49 (p = .01). However, the Tukey’s Test identified no individual interaction effects that were significant.

Error Perception Discussion

Our first research question examined which errors bothered professional communication practitioners and academics. The results provide a current data point that can be used to track how error perception has evolved over the last several decades.

The first five errors listed in Table 3 are ones this study’s participants ranked as the most bothersome (presented in descending order of statistical mean; see Table 1). Participants in this study also rated fancy tone/language, logic/sequence error, missing or needed space, and incorrect or missing question mark as the most bothersome errors; however, these previous studies did not analyze these error types. The additional errors listed are errors that were more bothersome to participants in previous studies or bothersome errors that were not assessed in this study. We present this data alongside the results of five previous taxonomies, which represent both practitioner (Beason, 2001; Gilsdorf & Leonard, 2001; Hairston, 1981) and academic perceptions (Kantz & Yates, 1994; Lunsford & Lunsford, 2008) as well as more than 35 years of related research.

Overall, data in Table 3 reflect previously noted observations about error research: a diversity of error types that were measured with a variety of methodological approaches in a variety of populations. Our results appear most similar to Lunsford and Lunsford (2008). This could be related to our methodologies, which were both informed by error data from our respective fields. Additionally, it is the most recent study, and our combined results could indicate changing perceptions of error in the past decade.

Table 3. This Study’s Most Bothersome Errors Compared With Five Previous Studies.

Error	This study	Lunsford and Lunsford (2008)	Gilsdorf and Leonard (2001)	Beason (2001)	Kantz and Yates (1994)	Hairston (1981)
Sentence fragment	Moderately bothersome (5/7 scale)	Ranked 17/20 frequently marked errors by teachers	Ranked 1 and 6/10 most distracting errors	Definitely bothersome (3/4 scale)	Ranked 4/11 most bothersome errors (4/6 scale)	Very serious (5/6 scale)
Unnecessary or missing apostrophe	Moderately bothersome (5/7 scale)	Ranked 9/20 marked errors	Ranked 5 and 8/10 most distracting errors	Somewhat bothersome (3/6)		Minor or unimportant (1/6 scale)
Misspelling	ᵃModerately bothersome (5/7 scale) and neutral/no opinion (4/7)	Ranked 2/20 marked errors		Somewhat bothersome (2/4 scale)	Homophonic misspellings ranked 2, 3, 7, and 8/11 most bothersome errors (4/6 scale)	Very serious (5/6 scale)
Wrong word	Moderately bothersome (5/7 scale)	Ranked 1/20 marked errors		Somewhat bothersome (3/6 scale)		Moderately serious (2/7 scale)
Poorly integrated source material	Moderately bothersome (5/7 scale)	ᵇRanked 3 and 19/20 marked errors
Unnecessary shift in verb tense	Neutral/no opinion (5/7 scale)	Ranked 11/20 marked errors		Somewhat bothersome (3/6 scale)		Serious (4/6 scale)
Unnecessary or missing capitalization	Neutral/no opinion (5/7 scale)	Ranked 7/20 marked errors				Very serious (5/6 scale)
Fused (run-on) sentence		Ranked 18/20 marked errors	Ranked 3/10 most distracting items	Somewhat bothersome (2/4 scale)	Ranked 11/11 most bothersome errors (4/6 scale)	Very serious (5/6 scale)

ᵃMost previous studies collapse misspellings into a single category. Data in these cells reflected results from the errors in this study labeled homophone and general misspellings. ᵇLunsford and Lunsford (2008) labeled these error types as incomplete or missing documentation and poorly integrated quotation.

Previous research also indicated that academics were typically more bothered by error than practitioners. However, the practitioners in this study were collectively more bothered by 19 of the 24 errors than academics (see Table 1). Seven statistically significant between-group differences were also identified, six of which bothered practitioners more than academics.

One explanation for the increase in practitioner botheration could relate to methodology. Previous studies provided little detail on why they tested specific errors, suggesting these errors were identified based on personal preference or anecdotal evidence. The errors assessed in this study were informed by an error analysis of editing tests, which hiring managers used to screen prospective technical communicators (Boettger, 2014). Other studies have similarly identified errors within existing texts instead of assessing error from a predefined list (Connors & Lunsford, 1988; Lunsford & Lunsford, 2008). By deriving errors from an authentic workplace genre, our results offer several errors not assessed in previous research—particularly those related to content, style, and design—which perhaps explains our practitioners’ botheration levels.

Furthermore, half of the statistically significant differences in error perception were related to design: space missing or needed, indentation error, and incorrect text styling. Each of these error types ranked in one of the three botheration tiers, and participants could typically describe or identify them correctly (see Table 2). These identification results suggest that the overall error rankings reflected participants’ true perceptions. Overall, errors related to formatting (e.g., missing spaces, incorrect paragraph alignment) are easier to identify than word- or sentence-level errors. In fact, many of these errors can be identified without reading any actual text. As shown in the appendix, the questions written for the alignment and indentation errors used Greeked text—placeholder text intended to focus readers’ attention on structure rather than content. Williams (1981) reminded us that we are often more bothered by errors that we know well or can find with ease. Greeking these questions could have primed participants to look specifically for design errors. Professional and technical communication practitioners were likely bothered by design errors because of their increased exposure to them. Writers and editors are often charged with creating company style guides and creating document templates, tasks that likely heighten their awareness to errors related to consistency and structure. Future research could examine if design errors are as bothersome to practitioners outside of professional communication.

Williams’s (1981) suggestion that we tend to be more bothered by errors that we know or can easily identify might also explain the rank of the remaining errors, particularly those in the most bothersome tier. For at least the past 40 years, fragments have consistently ranked as bothersome (Beason, 2001; Connors & Lunsford, 1988; Gilsdorf & Leonard, 2001; Hairston, 1981; Kantz & Yates, 1994; Lunsford & Lunsford, 2008). The severity of this error has likely been passed down to (and internalized by) many English writing and grammar students, perhaps explaining its top rank in this study as well as being the only error that academics were significantly more bothered by than practitioners. Furthermore, incorrect or missing question mark could have ranked higher than other punctuation-related errors like missing comma after an introductory element and hyphen, en-dash, and em-dash error because of placement. Question marks are placed at the end of a sentence and easier to spot than commas or hyphens, which are placed within words and clauses as well as contain alternate usage rules. Ease in identification might also explain why misspellings related to homophones were more bothersome (and more correctly identified) than general misspellings.

On the other hand, the rankings of these error types could also be an illustration of participants’ language preferences or character judgments. Beason (2001) observed both in the follow-up interviews with his participants. These participants often commented that fragments and fused sentences were marks of poor writing, giving the appearance that the original writers could not articulate complete thoughts. Brandenburg (2015) also found that writers’ credibility correlated with the errors in their writing. When considering our list of most bothersome errors, it could be argued, then, that fragments ranked first because they impair comprehension—readers must correct the error themselves to gain understanding. This rationale could also account for the higher bothersome ranks of fancy tone/language, wrong word, logic/sequence error, and poorly integrated source materials. In turn, this might also explain the rank of less bothersome errors, including hyphen errors, incorrect number format, and British misspellings. While still somewhat bothersome, these errors do not (comparatively) decrease cognition and recall.

Finally, previous studies have suggested that error perception has fluctuated and even decreased over the past 40 years. For example, Gray and Heuser (2003) reported their participants were less bothered by fragments and capitalization errors than Hairston’s (1981) participants. Conversely, Boettger (2014) more recently identified capitalization as the second most common error that hiring managers placed in editing tests. He linked the frequency of this error to hiring managers’ desire to assess how accurately their applicants could apply the capitalization standards outlined in style guides. Other studies, such as Lunsford and Lunsford (2008), have also reported decreases in certain error types, but they too attributed these changes to context. Wrong word errors replaced misspellings as the top error from their previous study (Connors & Lunsford, 1988), but the researchers attributed this shift to electronic spell-checkers; the technology helps students remedy misspellings, but a reliance on the automated spelling suggestions likely correlates to the increase in wrong words. All these results collectively bolster Hairston’s (1981) claim that not all errors are created equal. Our study reflects a fluidity in error botheration compared with earlier studies; however, we cannot claim that overall perception has decreased. The results that this and other studies have reported are context dependent; the overall perception of a single error must be considered in relation to the other errors assessed in that study.

Error Recognition Discussion

Our second research question examined how participants’ error perceptions correlated with their ability to recognize errors. Table 4 organizes the error types explored in this study by their correct identification range. Participants correctly identified or described 16 of the 24 errors 51% or more of the time. Correspondence analysis results suggested that seven of these error types corresponded with participants’ correct identification (Figure 1). Participants ranked four of these errors (sentence fragment, incorrect question mark, homophone misspellings, and wrong word; see Table 1) in the moderately bothersome tier, demonstrating that error botheration might correlate with correct identification. Participants ranked the remaining corresponding errors (indentation error, sentence ends with a preposition, and incorrect alignment) in the neutral/no option tier. These ranks also provide valuable information and indicate that participants recognized the errors but were neutral to their presence.

Table 4. Errors Organized by Identification Range and Botheration Rank Compared With Botheration Rank in Boettger (2014).

Error type	Broad error	Botheration rank (scale)	Rank in Boettger (2014)
Correctly identified ⩾80%
Unnecessary or missing apostrophe	Punctuation	2/24 errors (5/7)	13/71
Misspelling (homophone)	Spelling	3/24 (5/7)	ᵃ2/71
ᵇSentence fragment	Grammar	1/24 (5/7)	27/71
Incorrect or missing question mark	Punctuation	8/24 (5/7)	69/71
ᵇIndentation error	Design	11/24 (4/7)	69/71
Correctly identified 51%-79%
Incorrect alignment	Design	15/24 (5/7)	41/71
ᵇSpace missing or needed	Design	7/24 (5/7)	17/71
Sentence ends with preposition	Grammar	16/24 (4/7)	65/71
Wrong word	Content	5/24 (5/7)	1/71
Incorrect number format	Style	18/24 (4/7)	7/71
ᵇIncorrect text styling	Design	20/24 (3/7)	22/71
Fancy tone/language	Style	4/24	41/71
ᵇLogic/sequence error	Content	6/24 (5/7)	37/71
ᵇMissing comma after an intro element	Punctuation	17/24 (4/7)	17/71
Unnecessary or missing capitalization	Grammar	13/24 (4/7)	3/71
Unnecessary negative construction	Style	19/24 (3/7)	54/71
Correctly identified ⩽50%
Misspelling (general)	Spelling	10/24 (4/7)	ᵃ2/71
Hyphen, em-dash, or en-dash error	Punctuation	23/24 (3/7)	6/71
Vague or missing language	Content	14/24 (4/7)	17/71
ᵇMisspelling (British)	Spelling	24/24 (3/7)	ᵃ2/71
Redundant, expendable, or incomparable language	Style	22/24 (3/7)	4/71
Misspelling (proper noun)	Spelling	21/24 (3/7)	ᵃ2/71
Unnecessary shift in verb tense	Grammar	12/24 (4/7)	7/71
Poorly integrated source material	Content	9/24 (5/7)	62/71

ᵃIn this study, the researcher collapsed misspellings into a single category. Fifty-one percent of this sample included general misspellings, 16% homophone misspellings, 14% proper noun misspellings, and 6% British misspellings. ᵇSignificant at ⩽.05 alpha level between practitioners and academics. Practitioners were significantly more bothered by every error except fragments.

Data in Table 4 also show that eight error types were correctly identified 50% or less of the time. Three of these errors corresponded with incorrect identification: misspelling (proper noun); redundant, expendable, or incomparable language; and vague or missing language. Two of these errors appeared in the somewhat bothersome tier, and vague and missing language was ranked in the neutral/no opinion tier. Lack of context and prior knowledge (in the case of the proper noun misspellings) could have resulted in the frequent misidentification of these errors. Errors isolated in a single sentence may not be the optimal methodological design for assessing style errors like wordiness or vague language. And while proper nouns, such as a company’s name, are often purposely placed in employment screening tests for professional communicators (Boettger, 2014), our participants were not encouraged to use outside references or conduct research prior to completing the survey.

Similar to Brandenburg’s (2015) results, we found it challenging to code participants’ intent, particularly with errors that we classified as incorrectly identified. For example, when asked to identify unnecessary shifts in verb tense—which were correctly identified 28% of the time—participants often responded with anger or sarcasm. One academic wrote that the sentence included “ridiculous platitudes” and would earn a grade of D or D−. Similarly, a practitioner wrote that the sentence “was written by George Bush, no more need be said.” These types of responses make it difficult to determine if participants could not identify the error or if recognizing the error magnified a language bias. Beason (2001) noted similar types of reactions in his participant interviews, and Gray and Heuser (2003) wondered if inconsistencies in error recognition could lead to subordinates being judged unfairly in the workplace.

The results of this study (and others) have implied correlation between error perception and identification. However, an issue that also merits exploration is how these opinions reflect the types of errors actually found in specific contexts. For the purposes of this study, are professional and technical communicators bothered by the same errors that they frequently encounter in the workplace? A way to measure this is to first assess errors found in an authentic body of texts (e.g., Boettger, 2014; Connors & Lunsford, 1988; Lunsford & Lunsford, 2008). For example, Lunsford and Lunsford (2008) found that missing a comma after an introductory element was the second most frequent error in college student writing, but it was only marked by teachers 28% of the time. Findings like this suggest that teachers made judgment calls when marking errors and that the most frequently occurring errors did not always reflect their personal beliefs.

Comparable data sets are not as readily available in professional communication because of the privatization of most workplace writing. Boettger’s (2014) study of technical editing tests provides the most contextualized data on workplace error in the field. He hypothesized that the errors purposely placed in these tests reflected how important their identification was to hiring managers. The most significant result from this study was that 75% of the errors that practitioners found most bothersome did not correlate with the frequency or dispersion of those same errors in the editing tests. We identified similar trends in this study. The last two columns in Table 4 present the overall rank of the 24 errors assessed in this study compared with the weighted rank in Boettger’s (2014) study. Several of the errors that were infrequent and weakly dispersed in the previous study—sentence fragment, incorrect question mark, fancy tone/language, and space missing or needed—were errors the present study’s participants ranked most bothersome. Similarly, several of the errors typically found in editing tests—general misspelling; hyphen, em-dash, or en-dash error; redundant, expendable, or incomparable language; and unnecessary shift in verb tense—were also errors the present study’s participants could only identify 50% or less of the time. These disparities merit further exploration, particularly for determining which errors to emphasize in classroom teaching.

Social Variable Discussion

Our final research question examined if error perception interacted with a participant’s gender, job type, or years of work experience. Overall, results indicated that participants’ gender influenced error perception. Females were significantly more bothered by grammar, punctuation, spelling, and content errors than males. Previous studies reported that females also appeared more bothered by errors than males, but researchers attributed these differences to the gender distributions of their sample. These samples were typically dominated by female academics and male practitioners (Gilsdorf & Leonard, 2001; Hairston, 1981; Leonard & Gilsdorf, 1990). The resulting samples also informed the hypothesis that academics were significantly more bothered by error than practitioners. Our study also skewed female (more than 70% of the sample), but the females dominated both the practitioner and academic samples. Our sample demographics, then, could account for why gender appeared to influence perception more than job type or years working in the field. It is unclear if our sample’s demographics represent professional and technical communication (no empirical census of the field exists). However, the fact that substantially more females voluntarily participated in this survey could confirm their interest in usage error.

Various explanations have been offered on why females are more bothered by error than males. Researchers have posited that females are socially conditioned to care about writing errors, females are more in touch with their feelings than males, and males may be less willing to acknowledge being bothered by things like usage error (Kantz & Yates, 1994). However, we return to the empirical evidence that females typically choose more formal registers than males (Finegan & Biber, 2001; Fries, 1940), accounting for their botheration level. Furthermore, we recall the emotional and psychological properties that appear to inform error perception. Empirical evidence on at least 16 different language features suggests that females often use words related to psychological and social processes, but males refer more to objective properties and impersonal topics (Mulac, Bradac, & Gibbons, 2001; Newman, Groom, Handelman, & Pennebaker, 2008). It is possible, then, that females generally have a deeper connection to language and, therefore, react more strongly to language errors than males. There also may be additional social variables that interact with or better inform the gender significance, including participants’ native speaker status, ethnicity, the amount of writing they encounter daily, the types of classes they teach (e.g., lower level undergraduate, upper-level graduate), and the purpose of reading this writing (e.g., for grammatical correctness, for content). Our findings further support that context influences our general perceptions of errors, so future research must more thoroughly account for the demographics of the participants who are registering these perceptions.

Conclusion

We designed this study to extend research in error perception and recognition. Our results offer new error types for teachers and practitioners to consider, and we addressed the methodological issues of previous research. But this study has its limitations. We continued with the tradition of previous botheration studies by asking participants to respond to sentences that contained one isolated error. Though we varied and randomized our sentences, we still primed participants to look for error. Williams (1981) argued that the trouble with this approach—asking whether X is or is not a good usage—is that people are likely to answer: “Since we can ask an indefinite number of questions about an indefinite number of items of usage, we can, merely by asking, accumulate an indefinite number of errors, simply because whoever we ask will feel compelled to answer” (p. 154). Williams recommended that all error results be viewed with some skepticism and, in turn, future studies should adopt refined methodological designs.

Additional research should continue to measure error perception, experiment with new approaches for measuring error identification, and assess how these variables correlate. Furthermore, attention should be paid to how social variables influence error perception. To better understand why specific errors are bothersome, we must understand what factors influence these reactions. Previous research has focused almost exclusively on the errors being assessed rather than the participants who were assessing them. New studies should collect data on participants’ demographics as well as the contexts where these participants encounter error. This type of research complements the need to study error via authentic contexts and text types, an approach that is much needed in professional and technical communication.

Williams (1981) wrote at length about the variety of lists containing the “most bothersome” errors, and our results certainly add to this content variety. These inconsistencies across studies, however, provide little direction for how professional and technical communication teachers should prioritize error in their classrooms. We hope the addition of error types, particularly those related to content, style, and design, provides an alternative, more contextualized error list for classroom teaching. Related classroom instruction could focus on creating style sheets for technical documents and engaging students with the document styles tool in word processing programs, which can be used to reduce the presence of errors related to text formatting and styling. Students must also understand that error competency is often determined subjectively and applied inconsistently; however, professional situations still demand fluency in grammar, punctuation, and style. Minor errors in job materials, reports, and procedures can be costly and litigious and can directly impact the credibility of writers. In turn, teachers must also recognize that their personal beliefs do not always reflect the errors most common in the workplace. Exploration of error will continue to address the fluid concept of “correctness” and the factors that influence our own language preferences and biases.

Appendix

Survey Scale and Sentences

The survey asked participants to register their botheration level of 24 usage errors on a 7-point Likert-type scale.

1. Not bothersome
2. Slightly bothersome
3. Somewhat bothersome
4. Neutral/no opinion
5. Moderately bothersome
6. Very bothersome
7. Extremely bothersome

Practitioners were asked to consider each error as if they were evaluating the writing of a prospective employee. Academics were asked to consider each error as if they were evaluating the writing of a senior-level student.

The below questions were randomized; no participants saw the same question in the same order. Participants were also randomly shown one of two possible sentences for each of the 24 usage errors.

Grammar

1. Unnecessary or missing capitalization

   • Visitors Must Register All Cameras with the Attendant at the Entry Station.
   • The principal agency involved is the department of agriculture; however, the budget bureau is also peripherally concerned.

2. Unnecessary shift in verb tense

   • We take great pleasure in welcoming you to our staff. We hope our relationship is one of mutual understanding and support. The owners had many years of experience in the operation of successful and profitable businesses. We were fortunate in the past with our choices for our staff, and we sincerely hope that you will follow in this path.
   • After you press the Alt and P keys together, the computer will send your file to be printed.

3. Sentence fragment

   • The staff wants additional benefits. For example, the use of company cars.
   • Two years ago, a similar study was done by members of the accounting department. However, this study was negated. Because it was based on outdated estimates of the costs involved.

4. Sentence ends with preposition

   • He is the candidate I spoke of.
   • The branch office is where she was at.

Punctuation

5. Hyphen, em-dash, or en-dash error

   • We need a biological waste management system.
   • Using 12 point leading with a 10 point font opens up the text, making it look more inviting.

6. Unnecessary or missing apostrophe (including its/it’s)

   • Our firms performance has been excellent.
   • As McAuliffe pointed out, its a good thing that the auditors talked to the department before making the findings public.

7. Missing comma after an introductory element

   • Because hard-copy editing has ergonomic advantages Chico prefers it to onscreen editing.
   • In the appendix of our style guide we have listed the correct spellings for specialized terms that are frequently used in our publications.

8. Incorrect or missing question mark

   • Does anyone know when the copy is due back to layout???
   • He asked me where I put the specifications?

Spelling

9. General

   • Our analyses do not correlate with the preliminary test cores.
   • A former employee was charged with sexual harassment in an embarassing and costly lawsuit.

10. Homophone

    • Research has proven that Christmas bonuses do not effect the quality or quantity of work done by the recipients.
    • During the meeting, she sited data from the latest research study.

11. British

    • The Northeast suffered its worst storm of the season, and the government officials cancelled their office hours for the next two days.
    • Steffi used colour theory to design her Web site.

12. Proper nouns

    • Our company’s celebrity spokespeople include Sybil Shepard and Julia Louise-Dreyfus.
    • Since 1976, National Instrument has equipped engineers and scientists with tools that accelerate productivity, innovation, and discovery.

Content

13. Wrong word

    • The subcontracting was distributed between three firms.
    • I did not realize that half the accounting staff had a severe allegory to peanuts.

14. Vague or missing language

    • Place document in tray with printed side facing opposite.
    • The committee met for several hours and got a lot done.

15. Poorly integrated source material

    • A 1970s study of what makes food appetizing “Once it became apparent that the steak was actually blue and the fries were green,” some people became ill (Schlosser 565).
    • The Video Alert and Control (VAC) dashboard system has the “ability to visualize the road ahead and its dangers and to display this information graphically to the driver on a video screen on the dashboard.”

16. Logic/sequence error

    • To shut down the operating system, first exit all open programs and then save and close all open files.
    • We examined three storage methods most frequently used in our industry: (1) Trax, (2) Stacker, (3) Wide-Aisle Racking, and (4) Floor Storage.

Style

17. Redundant, expendable, or incomparable language

    • The cells were red in color.
    • Nevertheless, it is important to note that the monomers of proteins, lipids, and carbohydrates are neither energetically nor nutritionally equivalent, and that tracer molecules belonging to the same class may be allocated and used differently in the body.
    • Asia’s situation is somewhat unique, in that many of the culture ponds are owned and managed by small-scale farmers rather than by large corporations.

18. Incorrect number format

    • The purchase order was for 12 6-foot tables.
    • 15 of the study’s participants took 50 mg of caffeine and 110 ring of aspirin along with 25 mg of ephedrine twice daily.

19. Unnecessary negative construction

    • It is not without regret that I offer my resignation.
    • Never fail to back up your disks.

20. Fancy tone/language

    • I am in receipt of your memo requesting an increase in pay and am of the opinion that it is not merited at this time due to the fact that you have worked here for only one month.
    • Please utilize a canine feces collection instrument to optimize sanitary conditions within perimeters of the park grounds.

Design

21. Space missing or needed

    • Inadequate quality-control procedures have resulted in more serious equipment failures.Quality control is our most serious problem at present.
    • Inadequate quality-control procedures have resulted in more serious equipment failures. Quality control is our most serious problem at present.

22. Incorrect text styling

    • Greg has been promoted to Senior Vice President of Marketing. Greg’s replacement for General Manager will be announced next week.
    • Erin interviewed for the usability specialist position at The Home Depot.

23. Incorrect alignment

    Heading 1 Lorem ipsum dolor sit amet, porta et hymenaeos per, interdum ornare id facilisi lobortis nec. Senectus nunc vel urna, magna non, sed nullam nec eget mauris risus, a rhoncus.

    Heading 1 Lorem ipsum dolor sit amet, porta et hymenaeos per, interdum ornare id facilisi lobortis nec. Senectus nunc vel urna, magna non, sed nullam nec eget mauris risus, a rhoncus.

24. Indentation error

    Heading 1 Lorem ipsum dolor sit amet, porta et hymenaeos per, interdum ornare id facilisi lobortis nec. Senectus nunc vel urna, magna non, sed nullam nec eget mauris risus, a rhoncus.

    Heading 1 Lorem ipsum dolor sit amet, porta et hymenaeos per, interdum ornare id facilisi lobortis nec. Senectus nunc vel urna, magna non, sed nullam nec eget mauris risus, a rhoncus.

    Heading 1 Lorem ipsum dolor sit amet, porta et hymenaeos per, interdum ornare id facilisi lobortis nec. Senectus nunc vel urna, magna non, sed nullam nec eget mauris risus, a rhoncus.

Authors’ Note

This study was approved by the Institutional Review Board of the University of North Texas (Approval #12-320). Participant comments are reproduced by permission.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

1. Alred, G. J., Brusaw, C. T., & Oliu, W. E. (2012). The handbook of technical writing (10th ed.). Boston, MA: Bedford/St. Martin’s.
2. Amare, N., Nowlin, B., & Weber, J. H. (2011). Technical editing in the 21st century. Boston, MA: Prentice Hall.
3. Beason, L. (2001). Ethos and error: How business people react to errors. College Composition and Communication, 53, 33-64.
4. Boettger, R. K. (2011). Examining error in the technical communication editing test. In Proceedings of the Technical Communication Summit: STE’s 58th Annual Conference (pp. 142-147). Fairfax, VA: Society for Technical Communication. Retrieved from https://summit.stc.org/wp-content/uploads/2011/05/proceedings_2011-FINAL.pdf
5. Boettger, R. K. (2012). Types of errors used in medical editing tests. American Medical Writers Association Journal, 27, 99-105.
6. Boettger, R. K. (2014). The technical communication editing test: Three studies on this assessment type. Technical Communication, 61, 215-231.
7. Boettger, R. K., & Lam, C. (2013). An overview of experimental and quasi-experimental research in technical communication journals (1992-2011). IEEE Transactions on Professional Communication, 56, 272-293.
8. Brandenburg, L. C. (2015). Testing the recognition and perception of errors in context. Business and Professional Communication Quarterly, 78, 74-93.
9. Connors, R. J., & Lunsford, A. A. (1988). Frequency of formal errors in current college writing, or Ma and Pa Kettle do research. College Composition and Communication, 39, 395-409.
10. Finegan, E., & Biber, D. (2001). Register variation and social dialect variation: The register axiom. In P. Eckert & J. R. Rickford (Eds.), Style and sociolinguistic variation (pp. 235-267). Cambridge, England: Cambridge University Press.
11. Fries, C. C. (1940). American English grammar: The grammatical structure of present-day American English with especial reference to social differences or class dialects. New York, NY: Appleton-Century-Crofts.
12. Gilsdorf, J., & Leonard, D. (2001). Big stuff, little stuff: A decennial measurement of executives’ and academics’ reactions to questionable usage elements. Journal of Business Communication, 38, 439-471.
13. Gray, L. S., & Heuser, P. (2003). Nonacademic professionals’ perception of usage errors. Journal of Basic Writing, 22, 50-70.
14. Gurack, L. J., & Hocks, M. E. (2009). The technical communication handbook. New York, NY: Pearson Longman.
15. Hairston, M. (1981). Not all errors are created equal: Nonacademic readers in the professions respond to lapses in usage. College English, 43, 794-806.
16. Kantz, M., & Yates, R. (1994). Whose judgments? A survey of faculty responses to common and highly irritating writing errors. In Partial proceedings of the Annual Conference of the Assembly for the Teaching of English Grammar (pp. 38-54). Urbana, IL: National Council of Teachers of English. Retrieved from http://eric.ed.gov/?q=ED397435&id=ED397435
17. Lam, C. (2014, October). Visualizing categorical data: An introduction to correspondence analysis for technical communication researchers. Paper presented at the 2014 IEEE International Professional Communication Conference, Pittsburgh, PA. https://doi.org/10.1109/IPCC.2014.7020345
18. Leonard, D. J., & Gilsdorf, J. W. (1990). Language in change: Academics’ and executives’ perceptions of usage errors. Journal of Business Communication, 27, 137-158.
19. Locker, K., & Kienzler, D. (2012). Business and administrative communication (10th ed.). New York, NY: McGraw-Hill/Irwin.
20. Lunsford, A. A., & Lunsford, K. J. (2008). “Mistakes are a fact of life”: A national comparative study. College Composition and Communication, 59, 781-806.
21. Lunsford, A. A., Matsuda, P. K., & Tardy, C. M. (2010). The everyday writer with exercises with 2009 MLA and 2010 APA updates (4th ed.). Boston, MA: Bedford/St. Martin’s.
22. Mulac, A., Bradac, J. J., & Gibbons, P. (2001). Empirical support for the gender–as–culture hypothesis. Human Communication Research, 27, 121-152.
23. Newman, M. L., Groom, C. J., Handelman, L. D., & Pennebaker, J. W. (2008). Gender differences in language use: An analysis of 14,000 text samples. Discourse Processes, 45, 211-236.
24. Noguchi, R. R. (1991). Grammar and the teaching of writing: Limits and possibilities. Urbana, IL: National Council of Teachers of English.
25. Rew, L. J. (1999). Editing for writers. Boston, MA: Prentice Hall.
26. Rude, C. D., & Eaton, A. (2011). Technical editing (5th ed.). Boston, MA: Longman.
27. Smith, E. L., & Bernhardt, S. A. (1997). Writing at work: Professional writing skills for people on the job. Lincolnwood, IL: NTC.
28. Williams, J. M. (1981). The phenomenology of error. College Composition and Communication, 32, 152-168.
29. Wolfe, J., Shanmugaraj, N., & Sipe, J. (2016). Grammatical versus pragmatic error: Employer perceptions of nonnative and native English speakers. Business and Professional Communication Quarterly, 79, 397-415.

Author Biographies

Ryan K. Boettger is an associate professor and assistant chair in the Department of Technical Communication at the University of North Texas. His research areas include technical editing, content analysis, and STEM education. He is a former editor for the Texas Army National Guard.

Lindsay Emory Moore earned a PhD in English and a graduate certificate in teaching technical writing from the University of North Texas. She currently teaches technical writing within the Dallas–Fort Worth Metroplex.