Abstract
In public and academic debates, the gender star has been criticized for potentially hindering text comprehensibility and imposing an additional burden on learners of German by complicating an already complex language. We investigated in an experiment whether the use of the gender star (e.g., Student*innen) complicates text comprehensibility compared to masculine generics for learners of German. Participants were 80 Flemish students studying German as a foreign language. They were asked to read 13 short news articles, of which ten texts served as experimental items in either gender star or masculine form. Participants' accuracy in answering content questions, subjective ratings of text comprehensibility and sentence difficulty, and their full-text reading times were assessed. Additionally, we collected data on participants' proficiency in German, prior knowledge of, and attitudes toward gender-fair German. Our results suggest that the gender star does not pose a strong hindrance to text comprehensibility for Flemish students learning German: while proficiency in German had a significant effect on content question accuracy and sentence difficulty as well as subjective comprehensibility ratings, this effect was independent of gender form.
1 Introduction
Over the past few decades, gender-fair language (GFL) has become increasingly prominent in German-speaking countries. One of GFL's main goals is the avoidance of masculine generics, i.e., the use of masculine forms to refer to people of any gender when the referents' gender is unknown or irrelevant to the context, as this practice has been shown to render women and nonbinary people less visible in language (e.g., ; Zacharski and Ferstl, 2023). To counteract this bias, both neutralization and differentiation strategies have been proposed for the German language, with the latter including typographical innovations like the gender star (e.g., Student*innen) (Steinhauer and Diewald, 2017).
Despite the increasing adoption of GFL in institutional (Schneider, 2022) and professional contexts () and its empirically proven positive impact on mental gender representations and attitudes toward gender equality (e.g., ; ; ; Zacharski and Ferstl, 2023), GFL continues to be a subject of political debate and faces opposition within society (). Critics argue that innovative forms like the gender star create orthographic complexity and hinder text comprehensibility, particularly for learners of German (Rat für deutsche Rechtschreibung, 2021). However, empirical research with first language (L1) speakers suggests that gender-fair forms do not impair text comprehensibility (e.g., ; ), or that initial difficulties are overcome quickly (Zacharski et al., 2025).
This study seeks to address a notable research gap by experimentally investigating how GFL affects text comprehension for foreign language (LX) learners of German, more specifically by comparing texts that incorporate the gender star to those that use exclusively masculine forms. This will contribute to our understanding of how innovative GFL forms affect the language learning process in LX speakers. Building on the results from the present study, we aim to formulate recommendations based on empirical findings rather than subjective impressions and conjecture regarding the attention to and integration of GFL in the teaching of German as a Foreign Language (Deutsch als Fremdsprache or henceforth DaF).
2 Literature review
2.1 Gender-fair language in German
Languages differ on how deeply gender is ingrained in their grammatical structure (). German is a grammatical gender language. Hence, every noun carries a grammatical gender (masculine, feminine, neuter) to which noun-dependent forms such as articles and pronouns need to correspond. While the grammatical gender of inanimate nouns is arbitrary (e.g., der Löffel—the spoon [masc.], die Gabel—the fork [fem.], das Messer—the knife [neut.]), research has shown a considerable alignment between grammatical gender and the gender of the referent for animate words (e.g., die Mutter—the mother [fem.], der Onkel—the uncle [masc.]) (). In particular, for the majority of German role nouns, there are two forms: a masculine form (der Student—the (male) student [masc.]) and a feminine form (die Studentin—the female student [fem.]), which is derived from the masculine form by adding the feminine suffix -in [sing.]/-innen [plural]. However, even though both forms exist, there is an asymmetry in usage between feminine and masculine forms in German: while feminine forms may only be used to refer to women, masculine forms may be used generically, i.e., to refer to persons of any gender, whenever the referent's gender is unknown or irrelevant to the context.
In the context of the second feminist wave, the linguistic convention of using masculine generics was severely criticized alongside with other linguistic asymmetries, which in their combination highlighted the androcentric and sexist characteristics of German by degrading or making women less visible in language (e.g., Pusch, 1984; Trömel-Plötz, 1997). Indeed, although masculine generics are theoretically intended to include women aside from men, research has convincingly shown that they do so only to a limited extent (e.g., ). This process of problematizing these existing linguistic conventions went hand in hand with a movement toward GFL in German-speaking countries, a movement which gained momentum in the 1980s and 1990s. To avoid masculine generics, two strategies were proposed (e.g., ; Stahlberg et al., 2001): neutralization and differentiation. The neutralization strategy involves using gender-neutral terms that do not specify gender, such as Studierende [neut.] (those who study) instead of Studenten [masc.] (students). The differentiation strategy, on the other hand, involves explicitly mentioning different genders, such as using the pair form Studentinnen [fem.] und Studenten [masc.] (female students and male students) to explicitly denote both female and male students.
Initially, the movement toward GFL adopted a binary notion of gender. However, this binary perspective has evolved, especially since 2018, when the German Personal Status Act was amended to introduce a third gender option (called “divers”) for intersex individuals (). This legal change has increased the relevance of properly and accurately addressing and referring to people beyond the gender binary. To this end, the advantages of the neutralization strategy are increasingly highlighted, and advanced differentiation strategies have been introduced over the years through typographical innovations such as the gender star, which involves inserting an asterisk (*) between the masculine word stem (e.g., Student-) and the feminine suffix (-in [sing.] and -innen [plural]) (e.g., Student*innen) as an indication that all genders are included (Steinhauer and Diewald, 2017). Research has shown that using GFL is largely successful at making mental gender representations more balanced compared to masculine generics (e.g., ; ; ; ; Sato et al., 2016; Stahlberg and Sczesny, 2001). For the gender star in particular, it was found that its use can successfully overturn a male bias (; ; ; Schunack and Binanzer, 2022) and evoke inclusive mental representations that encompass nonbinary individuals, women, and men (Zacharski and Ferstl, 2023).
The use of GFL has already become widespread in German-speaking countries, especially in professional contexts. A clear diachronic evolution toward fewer masculine generics and more gender-inclusive forms was found in legal texts (), scientific texts (Toonen, 2022), newspaper reporting (Waldendorf, 2024), public speeches by German chancellors and presidents (), and debates and speeches in the German Bundestag (Stecker et al., 2021). In addition, texts produced by universities have been found to favor GFL (; ), and many companies are increasingly adopting GFL as part of their diversity efforts and commitment to non-discriminatory practices (). Findings indicate that this language change is mainly driven by progressive, left-leaning forces in society (Stecker et al., 2021; Waldendorf, 2024; ), and that the most frequently used gender-inclusive forms are pair forms and gender-neutral terms, with forms that contain typographical innovations such as the gender star occurring more rarely (; ; ). Moreover, proportions of masculine generics and GFL differ strongly depending on the context. While masculine generics were not used once in former German chancellor Olaf Scholz's first New Year's address () and are very rare in texts written by university administrations (; ), they still dominate in newspapers with an average between approx. 50 and 85 % of all role nouns (Waldendorf, 2024; ). In light of the overall increased use of GFL, many institutions and organizations have issued guidelines on the topic, further embedding these practices into society at large ().
While these developments indicate that language change driven by societal shifts is currently taking place, GFL still faces societal resistance and remains a topic of political controversy. Although progressive political parties in the German-speaking region tend to support GFL, conservative parties rather position themselves as opponents, criticizing what they perceive as an unnecessary linguistic intervention and a limitation to free speech. Additionally, two recent representative surveys, one conducted by the German public broadcasting institution Westdeutscher Rundfunk (WDR) [West German Broadcasting] in September 2022 and another by YouGov in March 2023, indicated that a majority of 62% and 69% of Germans, respectively, consider GFL to be unimportant or less important. However, the WDR survey brings some nuance to this observation, as it also showed that pair forms and neutralizations are already widely accepted and that it is mainly the forms with typographical innovations, i.e., the use of special characters such as the gender star in a personal noun before the female ending, that people take issue with (Westdeutscher Rundfunk and infratest dimap, 2022; YouGov, 2023; see also ; ; Zacharski, 2024). Further signs of the contested status of these special characters include the Council for German Orthography opposing their inclusion in the official rulebook (Rat für deutsche Rechtschreibung, 2021), and the fact that the German federal states of Bavaria, Saxony, Saxony-Anhalt, and Hesse have forbidden their use in their administration and/or schools. Typical arguments against the innovative, so called “non-orthographic” forms are that they create too much orthographic complexity and confusion and that they distract from the essential content of a text or make texts harder to read, especially for children, learners, and the visually impaired (e.g., ). At the same time, some defend these innovative forms. As () reasons, much depends on how one interprets “correct language use”: using these forms may violate existing orthographic rules, but they can be the only available option to correctly and accurately refer to or include people who identify beyond the binary.
Of all the innovative, non-orthographic forms, the gender star is arguably the most popular one (; ), also in a German as a Foreign Language (DaF) context (Stark, 2022), and its use is on the rise (). Some universities explicitly recommend the use of the gender star in their GFL guidelines (Schneider, 2022), and the outcome of a referendum in Zürich, Switzerland, on the gender star in November 2024 was positive, with a majority of voters in favor of the city continuing to use the symbol in its communication.
2.2 Gender-fair language in the context of German as a foreign language
Against this background, the present study aims to focus on the gender star in a DaF context. While some argue that the gender star poses a hindrance for learners of German (), others (; Stark, 2022; ; ; Stummer, 2024; Wooten, 2025; ; ) believe it is important to make learners aware of its existence and address it in DaF materials and courses. Not doing so could otherwise result in comprehension difficulties. Moreover, it would disconnect learners from ongoing societal debates in the German-speaking world and make them incapable of referring to or including nonbinary persons in German. Additionally, it would not properly prepare them for future jobs involving teaching German or communicating professionally in German.
Previous research on GFL in a DaF context is sparse. Available studies primarily focus on how DaF learners and teachers evaluate GFL in German (; Stark, 2022; Truan, 2023; ), on the insufficient consideration given to GFL in DaF textbooks and didactics (Stark, 2022; Stummer, 2024; but see ), or on how to integrate and teach GFL in the DaF classroom (; ; Wooten, 2025; ; ). Truan (2023), for example, found that second language (L2) speakers were more open toward GFL in their L2 compared to their L1 and did not appear to share the emotional reactions linked to negative aesthetic evaluations of new and unfamiliar forms. With regard to GFL in DaF teaching resources, Stummer (2024) found that, despite an increase in the use of neutralization strategies, generic masculine forms still predominate, while typographical innovations are entirely avoided. Moreover, she observed a lack of metalinguistic explanations on how to form or use GFL in German. To date, no published studies have investigated how DaF learners actively produce gender inclusive forms in German, although we are aware of ongoing projects that aim to address this gap.
, , and Zacharski et al. (2025) explicitly call for research that examines how learners of German process gender-fair forms and tests the frequently stated objection that the introduction of new forms would further complicate the already complex German language, by considering the learners' perspectives. In response to this call, the present study examines if and to what extent the gender star makes texts less comprehensible compared to masculine generics for learners of German.
2.3 Comprehensibility of gender-fair forms
Text comprehension involves an interplay of word recognition, syntactic parsing and higher-level processes, such as retrieving world knowledge and drawing inferences (). Comprehension is successful when the combination of these processes results in a mental representation, the situation model, that reflects the content of the text and the writer's intentions. The comprehensibility of a text thus depends on many text features, including the exact wording of the text and the choice of particular lexical items, but also its structure or coherence (). The term readability puts the focus more on word and sentence level features of texts. In the context of evaluating the effects of various forms of gender-inclusive wording on reading and comprehension, the two terms have been used interchangeably and for the present study, we also regard them as synonymous.
Two classes of methods have been used to study comprehensibility. One type of method is to ask readers for their subjective evaluation of the text as a whole or of specific aspects (e.g., syntactic complexity, word difficulty; ). Another type of method are behavioral assessments, such as the performance on comprehension questions or memory tasks (for comprehensibility), or the measurement of reading times, eye movements or lexical decision times (for readability).
While the comprehensibility of gender-fair forms has not yet been tested for learners of German, scholars have already addressed this question for L1 German speakers. Results from several experimental studies show that gender-fair forms, which align with current orthographic rules such as neutralizations and pair forms, do not impair text comprehensibility (e.g., ; ; ). Steiger-Loerbroks and von Stockhausen (2014) found that gender-neutral forms required more processing effort than masculine forms, but this was only the case at early reading stages, indicating a rapid habituation effect.
Results are less clear-cut for binary gender-fair forms that are considered in violation of German orthography. For texts using capital-I forms (e.g., FlugbegleiterIn—stewardEss), reported no difference in subjective readability but noted longer full-text reading times compared to texts using orthographic forms (masculine forms, pair forms, and neutralizations). For texts using slash forms (e.g., der/die Lehrer/in—the [masc.]/the [fem.] male teacher [masc.]/female teacher [fem. suffix]), () found no difference in content question accuracy, subjective comprehensibility, and aesthetic appeal, but significantly lower subjective readability ratings compared to texts using masculine forms. They suggest that the more complex shape and lower familiarity with the slash form may explain these results.
approached the question of the readability of gender-fair texts from a different methodological angle. They analyzed, based on a large corpus, how much textual material would need to be changed if non-gender-inclusive German texts were rewritten to be gender-inclusive. They found that only 0.95% of all tokens would be affected by gender-inclusive language, suggesting that gender-inclusive texts are very similar to non-gender-inclusive ones. This calls “into question whether gender-inclusive German presents a substantial barrier to understanding and learning the language, particularly when we take into account the potential complexities of interpreting masculine generics.” (, p. 10).
So far, four studies have specifically examined the comprehensibility of the gender star as a non-orthographic gender-fair form that is intended to include all genders in a sample of L1 speakers (; ; Zacharski et al., 2024, 2025). Both and ( compared gender-inclusive texts with role nouns in the gender star form to texts with masculine generics on the measures of subjective comprehensibility, word difficulty, sentence difficulty and aesthetic appeal, which are part of the validated comprehensibility questionnaire of . In a first experiment, which used a text with mainly plural and only a few singular star forms (e.g., die Spieler*innen—the players), found that the gender-fair text with the gender star did not impair text comprehensibility but even positively affected it. However, in a second experiment using a text with more singular than plural star forms (e.g., der*die Spieler*in—the[masc.]*the[fem.] player[star]), the results were reversed, with the gender-fair text with the star form negatively affecting subjective comprehensibility, sentence difficulty, and aesthetic appeal (). A possible explanation for these differing results is the higher complexity of singular star forms due to the duplication of the article (der*die—the[masc.]*the[fem.]) required for them; this is not necessary for plural star forms, as there is only one definite plural article in German (die), which is used for both feminine and masculine nouns. It should be noted, however, that such complex singular star forms are not frequently used (Zacharski et al., 2025). In a similar experiment that used a text with mostly plural and just a few singular star forms, found no statistically significant difference in comprehensibility ratings between participants who read a gender-fair text and those who read a text using only masculine forms. They also considered interindividual differences and found that neither attitudes toward GFL nor the (non-)academic background of participants affected the assessment of the text written in GFL using the gender star.
While and both used subjective measures to assess the comprehensibility of the gender star, Zacharski et al. (2025) captured its readability through a lexical decision task as an implicit measure, based on the assumption that word recognition is a crucial component of the reading process. Presenting data from two experiments—one with a student sample (18–30 years) and one with a more heterogeneous group of older non-students (30–80 years)—their findings show that less than 3% of all participants (both students and non-students) rejected star nouns as German words. Contrary to their expectations based on the length and lower frequency of star words, reading time data showed that students accepted star nouns as quickly and frequently as feminine and masculine forms. Non-students accepted star nouns more slowly and less often, but their initial difficulties in lexical access, reflected in slower reaction times, were quickly overcome during the experiment. This suggests that the readability of the gender star is a matter of familiarity and practice. Additionally, when examining how interindividual differences might influence the ease of recognition, they found that the (non-)student status and age of participants predicted the ease of lexical access to star nouns—likely due to the students' higher familiarity with star words—while attitudes toward GFL did not.
Zacharski et al. (2024) further investigated the readability of the gender star using implicit measures by moving from the word level to the text level. In an eye-tracking study, they compared newspaper articles featuring the gender star to those using masculine and capital-I forms. Each student participant read 32 newspaper articles, with experimental items containing four instances of a stereotypically neutral role noun in the plural and same gender form. They found that—at least for students—the gender star form does not impede text readability; in fact, it appears to make texts even easier to read. Overall, available studies on the gender star suggest that, as long as combined articles in the singular form are avoided, the criticism that gender star words are too complex and make texts more difficult to read and understand is not warranted, at least not for L1 speakers of German.
As mentioned above, the comprehensibility of GFL for learners of German has not yet been investigated. However, a study of Speyer and Schleef (2019) on English as a L2 examined whether singular they as a gender-neutral pronoun presents a processing challenge for learners of English, also depending on their proficiency in L2 English. They conducted a self-paced reading study with students from Austria as participants, and found that advanced L2 speakers (B2-level or higher) performed similarly to L1 speakers (), showing no difficulty in processing singular they compared to gendered pronouns (he or she). In other words, their study did not support the argument that GFL in a L2—in this case English—hinders the language learning process.
3 The present study
To our knowledge, this study is the first to experimentally examine the comprehensibility of GFL in learners of German as their LX. Specifically, our study focuses on texts that use the gender star compared to those that use only masculine forms. Previous experimental research with L1 speakers has found that gender star words, especially when primarily used in the plural, do not impair text comprehensibility (; ). Can we expect results for learners of German to differ from those for L1 speakers? This may be the case: Zacharski et al. (2025) found slower reaction times for the recognition of star words among non-students compared to students, although they also observed a rapid habituation effect. They attributed these results to non-students being less familiar with the gender star. Similarly, learners of German are likely less familiar with gender star forms compared to L1 speakers, given that they have learned the language later in life and that DaF courses have been found to address the topic of GFL only marginally (e.g., ; Stark, 2022). Lower levels of familiarity with the gender star may be even more pronounced among learners of German whose L1 does not include similar gender-fair forms, as is the case in the present study, where the vast majority of participants are L1 speakers of Belgian Dutch. Although Dutch is a grammatical gender language and historically had a three-gender system similar to German, masculine and feminine genders have gradually lost their distinct markers and become formally indistinct. Moreover, Dutch differs considerably from German in the feminization of personal nouns, with the derivation of feminine forms from masculine ones often being absent or becoming increasingly unproductive, resulting in a preference for neutralization over differentiation as a GFL strategy (Van Hoof and Decock, 2026). Based on this reasoning, we formulate our first hypothesis (H1) as follows: For LX learners of German (with Dutch as their L1), texts with the gender star will negatively affect accuracy in answering content questions (H1a), subjective comprehensibility (H1b), perceived sentence difficulty (H1c), and full-text reading times (H1d) compared to texts with masculine forms.
There may also be interindividual differences among learners depending on their level of proficiency, which can also be seen as an indication of their prior knowledge of or the amount of prior exposure to the gender star as a GFL form. Speyer and Schleef (2019) did not observe a processing challenge linked to singular they for advanced learners of English, but results may have been different for less proficient learners. Therefore, we hypothesize as follows: We expect lower proficiency in German to negatively affect text comprehensibility measures (content question accuracy, subjective comprehensibility, perceived sentence difficulty, full-text reading times) compared to higher proficiency (H2a), and we expect this effect of proficiency to be stronger for texts with the gender star than for texts with masculine forms (H2b).
Another individual variable often considered in studies on the comprehensibility of GFL is people's attitudes toward GFL. While Steiger-Loerbroks and von Stockhausen (2014) found that a more positive attitude toward the use of generic masculine role nouns correlated with longer first-pass reading times for GFL forms, the studies by and Zacharski et al. (2025) did not find a correlation between attitudes toward GFL and the comprehensibility of gender star forms. Given the contradictory research findings on this issue, our study aims to investigate how attitudes toward GFL relate to the comprehensibility of texts with the gender star form exploratively, without formulating a hypothesis.
4 Materials and methods
To study the effect of the gender star on text comprehensibility in learners of German, we conducted an online survey experiment. We received positive ethical approval for this study from the Faculty of Arts and Humanities at Ghent University. All materials and measures can be found in the accompanying repository on OSF (Zacharski et al., 2026).
4.1 Participants
The sample for the study consisted of Dutch-speaking learners of German. Our inclusion criteria required that participants be studying German as their LX at a higher education institution in the Flemish region of Belgium and that they had been doing so for at least one semester. Recruitment took place at two Flemish universities and two Flemish universities of applied sciences, all of which offer DaF courses at various levels and as part of different academic programs. Out of the 160 participants who initially started the experiment, 73 were excluded for not producing a full dataset (errors occurred during data collections: 36, early drop-out: 37), and seven were excluded for other reasons (no informed consent: 3, very low performance in control questions: 1, very fast reading times: 2, German as L1: 1). Our final sample thus consisted of 80 students (M = 19.48 years, SD = 2.29, range = 17–35; nonbinary: 1, female: 53, male: 26; L1 Dutch: 72, another L1: 3, L1 Dutch + another L1: 5), with 40 participants studying at a University (nonbinary: 1, female: 27, male: 12) and 40 participants studying at a University of Applied Sciences (female: 26, male: 14). Fifty-eight students reported having studied German in high school, while 22 reported that they had not.
An a priori power analysis was conducted using the simr-package in R (). Simulations (n = 1,000) based on the full model structure for sentence difficulty ratings indicated that a sample of 80 participants provides a power of 1—β = 81.3 % [95% CI: 78.7; 83.7] (α = 0.05) to detect small-to-medium effects of our primary variable of interest Gender_Form (b = 0.35, corresponding to Cohen's d = 0.35 given σ = 1) (see supplementary materials on OSF for a visualization of the power increase as a function of sample size; Zacharski et al., 2026).
4.2 Materials
For our experiment, participants were asked to read short newspaper articles. We used a selection of newspaper articles developed for the eye-tracking study by Zacharski et al. (2024). Newspaper articles are recommended and frequently used as didactic materials in more advanced DaF courses (from B2 level for receptive skills) (). To minimize respondent fatigue and avoid texts perceived to be higher in difficulty, two DaF lecturers who were familiar with the level of the students selected 13 newspaper articles out of the 32 used by Zacharski et al. (2024), aiming for texts on a B2 level. To verify whether the texts indeed covered a B2 level, the tool Wortliga (https://wortliga.de/textanalyse/api.php) was used. According to the tool, nine out of 13 articles had a B2 level, three were marked as C1, and one as C2. It was decided to include those texts as well, as they were not considered too difficult because of the close linguistic distance between German and Dutch. The newspaper articles cover a range of topics and average around 120 words each. Ten articles served as experimental items. For these items, gender forms were varied within subjects (gender star/masculine), ensuring that each participant read five texts with plural gender star role nouns (e.g., Kinobesucher*innen; see Figure 1) and five texts with plural masculine role nouns (e.g., Kinobesucher). Each role noun appeared four times per text in the same gender form, and each article is accompanied by a unique logo representing an imaginary regional newspaper or blog, designed to match the content. This design choice helps distract participants from noticing the manipulations of gender form in the texts, as it is common for newspapers to vary in their approach to GFL. To minimize unwanted text effects, two counterbalanced lists were created so that all texts were presented in both gender forms. To avoid potential interaction effects between gender form and gender stereotype (e.g., ; ; Sato et al., 2016; Vervecken et al., 2013), only stereotypically neutral role nouns were used in the experiment (). In addition, three newspaper articles served as filler items to further obscure the experiment's purpose. These filler texts differed from the other texts in that they did not contain inflectable role nouns; instead, they featured epicene and neutral forms. We pseudorandomized the order in which texts appeared to students to avoid participants seeing three consecutive filler texts, texts with gender stars, or texts with generic masculine forms.
Figure 1
4.3 Measures
Text comprehensibility was measured through performance on content questions, ratings on subjective comprehensibility and sentence difficulty, and full-text reading times. We assessed content question accuracy as an objective measure of text comprehension by asking one content question after each text, using a multiple-choice format with three response options, one of which was correct. We asked the same content question for each text as Zacharski et al. (2024). In addition, participants rated six items after each text, based on two subscales from the validated comprehensibility questionnaire by , which has also been used in subsequent studies investigating the effect of gender-fair forms on text comprehensibility (e.g., ; ; ): subjective comprehensibility (three items; e.g., All in all, the text was easy to understand) and sentence difficulty, which is defined as the perceived ease with which readers can analyze the syntax of the sentences of a text (three items, e.g., Many sentences were very long). These subscales were selected to keep the questionnaire concise and because they directly assess text comprehensibility and/or have been shown to be sensitive to gender star forms, as demonstrated in . All items were rated on 5-point Likert scales, ranging from 0 = I disagree to 5 = I agree. Two items were reverse-coded so that, overall, a higher score suggested more difficulties with text comprehension. With internal consistencies of Cronbach's α = 0.88 and α = 0.85, for sentence difficulty and subjective comprehensibility, respectively, both scales proved to be good. In addition, participants were asked to motivate their subjective evaluations in a text box. Furthermore, full-text reading times were recorded.
To assess students' proficiency in German and thereby account for interindividual differences, we relied on the LexTALE test. The test was originally developed for English, and a Dutch and German version were developed in parallel to the English version (). The test, which is a lexical decision task, has shown robust correlations with measures of general language proficiency. It consists of 60 vocabulary items (20 non-words and 40 words), where participants have to decide whether each item is a real word or not. Especially when time is short for data collection, the LexTALE is a suitable instrument to operationalize language dominance (e.g., ). It can be used to discriminate between lower intermediate or lower (below 59%), upper intermediate (60%−80%) and advanced (81%−100%) proficiency levels ().
Two additional measures relevant for the analysis of our data are participants' familiarity with the gender star and their attitude toward GFL. Familiarity with the gender star was coded as a binary variable (0: participants were not familiar with the gender star before taking part in the study; 1: participants have already encountered the form earlier on). Attitudes were measured using eight items from Zacharski (2024) testing general beliefs toward GFL in German (six items; e.g., I can understand why people with a nonbinary gender identity feel frustrated in a binary society) and recognition of the discriminatory power of binary gender forms toward persons identifying as nonbinary (two items; e.g., Do you find the following formulation discriminatory toward nonbinary individuals?: “Lieber Student, liebe Studentin, willkommen an unserer Universität!”). All items were rated on Likert scales ranging from 1 to 5. Four items were reverse-coded, so that a higher score suggest a more positive attitude toward GFL. The mean score of all eight items was used for data analysis. With an internal consistency of Cronbach's α = 0.76, the scale proved to be good.
4.4 Procedure and questionnaire
The experiment was integrated into participants' existing German courses at each institution and was considered an exercise in practicing reading and comprehension skills. This approach created a natural classroom setting and a realistic reading and learning environment. Participants did not receive any compensation for their participation. Data were collected in March 2024, and the survey was administered via Qualtrics. Students accessed the questionnaire during class via a provided link. At the beginning of the survey, they were informed that they were being asked to take part in a study on the comprehension of newspaper articles in German. They received information about the content and duration of the survey and were told that the survey was voluntary and would not affect their grades for the course. They were also informed that it was important to read each text carefully and that the test itself would not be graded. After this information, they were asked to give informed consent. Those who did not provide informed consent were given an alternative but similar task. Those who consented were then tasked with reading 13 newspaper articles, each followed by a content question, six items to rate and the option to justify their ratings. After reading the articles, students completed the LexTALE test, followed by a question about the study's intent. Based on their responses, it was clear that only one student recognized that the study was about the comprehensibility of the gender star.
The questionnaire continued with questions designed to gather demographic data (age, gender identity, first language(s), study program, study progress) and attitudinal data. It also included questions about whether students had studied German in high school and their familiarity with gender-fair language (GFL) in German. The survey ended with a thank you note for their participation and a debriefing about the true purpose of the research.
4.5 Data analysis
The data analysis was conducted with R (version 4.3.1) run on R studio (; version 2023.06.0). Attitudes toward GFL, familiarity with the star form, and LexTALE scores were analyzed using standard statistical methods, which will be reported in the results section. For the analysis of content question accuracy (800 datapoints), we fit a generalized mixed effect model (model M 1) using the glmer-function () of the lme4-package with the binary coded dependent variable response type for all non-filler texts (1: correct response, 0: incorrect response). Based on our hypotheses, we chose a three-way interaction term for fixed effects including the factorial variable Gender Form as well as the scaled covariates LexTALE and Attitude. Familiarity with the gender star was dropped to avoid multicollinearity of the model based on significant correlations between LexTALE scores and familiarity ratings as well as between attitude scores and familiarity which will be reported below. A parsimonious random effect structure was used to account for the experimental design (), thus including interindividual differences in response behavior to different gender forms (1+Gender Form|Participant ID) and potential item effects for the different texts used (1|Item). In order to guarantee model convergence, the allFit()-function of the lme4-package was used to identify bobyqa as model optimizer.
For the analysis of sentence difficulty ratings (2,400 datapoints), we fit a linear mixed effect model (model M 2.1) using the lmer()-function () of the lme4-package (version 1.1-33). The scaled rating scores (Likert-scale score 1–5) for all non-filler texts were used as the dependent variable. The fixed effect structure was the same as for the model of content question accuracy. For random effects, the additional term (1|Participant ID:Item) was added to account for the repeated measures design (three scale per text). A linear mixed effect model with the same fixed effect structure was fitted for the analysis of subjective comprehensibility ratings (2,400 datapoints, model M 2.2). However, the random slope for Gender Form was removed to guarantee model convergence (), thus resulting in the following random effect structure: (1|Participant_ID) + (1|Item) + (1|Participant_ID:Item). The default optimizer was used for both lmer-models of comprehensibility ratings.
Additionally, participants' comments in which they justified their text comprehensibility ratings were manually checked. We were particularly interested in explicit comments about the gender star and its impact on how they experienced reading the text. This allowed us to see if students related their ratings to the gender star.
Finally, before the statistical analysis of total text reading times for all non-filler texts, an outlier analysis was performed. First, from 800 datapoints in total, all trials with a total text reading time of <20 s (27 datapoints) and a total reading time of >200 s (eight datapoints) were excluded, in order to guarantee that participants performed the trial properly. Second, datapoints that were 2.5 SDs slower than the participant's mean response time were excluded (six datapoints). For the analysis of the remaining datapoints for total text reading times (759 datapoints), we fit a linear mixed effect model (model M3) using the lmer()-function () of the lme4-package (version 1.1-33). Logarithmized total text reading times were used as the dependent variable. The fixed effect structure remained the same as in the models previously described. However, in order to guarantee model convergence, we used the more modest random effect structure (1|Participant_ID)+(1|Item) ().
Using the rePCA()-function of the lme4-package showed that, for all models, the dimensions chosen for random effects accounted for 100% of the variability explained.
5 Results
5.1 Attitudes toward gender-fair language and familiarity with the gender star
A mean attitude score of M = 3.06 (SD = 0.70) suggested that participants of the study held, overall, a moderately positive attitude toward GFL (range = 1.44–4.78; scale: 1[negative] to 5[positive]). No significant differences in attitudes were found between female (M = 3.08, SD = 0.62, range = 1.78–4.67) and male participants (M = 2.96, SD = 0.78, range 1.44–4.56) (Wilcoxon rank sum test: W = 659, n1 = 26, n2 = 53, p = 0.758). Note, however, that the sample was not well-balanced with regard to binary gender identities (female: 53, male: 26).
Thirty-seven participants (nonbinary: 1, female: 24, male: 12) stated to have been familiar with this gender form before they took part in the study, while 43 participants (female: 29, male: 14) had not encountered it before. Calculating a simple binary logistic regression with the binary variable familiarity as the dependent and participants' binary gender identity as the categorical independent variable showed no significant effect of gender identity on the familiarity with the gender star form (Estimate: −0.01, CI: −0.25–0.23, p = 0.943).
A significant weak positive correlation was found between attitude scores and familiarity with the gender star form: participants with a more positive attitude toward GFL were more likely to have encountered the gender star form before they took part in the study [rpb(78) = 0.24, p = 0.03].
5.2 Proficiency in German: LexTALE scores
Overall, participants of our sample reached a mean LexTALE score of M = 61.98 (SD = 7.56). However, the wide range of 42.5–80 suggests large differences in proficiency between participants. We calculated a two-way ANOVA to test whether LexTALE scores were dependent on whether the participant has already studied German in school (1: yes, 0: no) or the type of university participants studied at (university vs. university of applied sciences; Table 1). While experience from school did not have an effect on LexTALE scores, the analysis yielded a significant effect of university type: participants studying at universities (M = 66.1, SD = 6.52, range: 51.2–80.0) reached significantly higher proficiency scores (i.e., upper intermediate level) than students at universities of applied sciences (M = 57.9, SD = 6.27, range: 42.5–72.5) (i.e., lower intermediate or lower proficiency level).
Table 1
| Predictor | Df | Sum Sq | Mean Sq | F-value | p-value |
|---|---|---|---|---|---|
| University type | 1 | 1,351 | 1,351 | 32.270 | <0.001*** |
| German classes in school | 1 | 1 | 1.4 | 0.034 | 0.853 |
| University type × German classes in school | 1 | 12 | 11.8 | 0.282 | 0.597 |
| Residuals | 76 | 3,182 | 41.9 |
Two-way ANOVA results for LexTALE-scores by university type and German classes in school.
Signif. codes: *** <0.001.
In order to test for potential influences of binary gender identity, an additional three-way ANOVA with the additional independent categorical variable gender identity (female vs. male; Table 2) was conducted: independent of university type, female participants (M = 61.1, SD = 7.69, range 42.5–80) reached significantly lower LexTALE scores than male participants (M = 63.9, SD = 7.24, range: 50–80). Again, however, this finding has to be treated cautiously, as our sample is not well-balanced with regard to binary gender identities of participants.
Table 2
| Predictor | Df | Sum Sq | Mean Sq | F-value | p-value |
|---|---|---|---|---|---|
| University type | 1 | 1,406.3 | 1,406.3 | 35,404 | <0.001*** |
| German classes in school | 1 | 1.6 | 1.6 | 0.040 | 0.841 |
| Gender identity | 1 | 184.2 | 184.2 | 4.638 | 0.035* |
| University type × German classes in school | 1 | 6.2 | 6.2 | 0.157 | 0.694 |
| University type × gender identity | 1 | 2.1 | 2.1 | 0.052 | 0.820 |
| German classes in school × gender identity | 1 | 65.1 | 65.1 | 1.638 | 0.205 |
| Residuals | 72 | 2,860 | 39.7 |
Three-way ANOVA results for LexTALE-scores by university type, German classes in school, and participants' binary gender identity.
Signif. codes: *** <0.001, * <0.05.
Finally, a moderate significant positive correlation was found between LexTALE performance and familiarity with the gender star form: participants with higher German proficiency were more likely to have encountered the gender star form before they took part in the study [rpb(78) = 0.417, p = <0.001]. In contrast, there was no significant correlation between LexTALE scores and attitude scores [rs(78) = 0.114, p = 0.316].
5.3 Text comprehensibility
5.3.1 Content question accuracy
On average, participants answered M = 78.0% (SD = 16.1, range: 30.0–100.0) of the content questions correctly. The analysis of content question accuracy (Table 3, Model 1) yielded a significant main effect of LexTALE scores: independent of gender form and attitudes toward GFL, a higher proficiency in German led to a higher proportion of correct responses in content questions (Figure 2).
Table 3
| Predictors | M1 content question responses | ||
|---|---|---|---|
| Odds ratios | CI | p-Value | |
| (Intercept) | 5.62 | 2.30–13.71 | <0.001*** |
| Gender form | 0.87 | 0.56–1.36 | 0.535 |
| LexTALE | 1.58 | 1.14–2.18 | 0.006** |
| Attitude GFL | 1.25 | 0.89–1.74 | 0.195 |
| Gender form × LexTALE | 0.97 | 0.65–1.44 | 0.883 |
| Gender form × attitude GFL | 0.83 | 0.55–1.25 | 0.367 |
| LexTALE × attitude GFL | 0.98 | 0.71–1.35 | 0.901 |
| 15.6-7.4,-14.3242pt(Gender form × LexTALE) × attitude GFL | 1.14 | 0.77–1.70 | 0.517 |
| Random effects | |||
| σ2 | 3.29 | ||
| τ00participantID | 0.44 | ||
| τ00item | 1.71 | ||
| τ11participantID × genderform | 0.00 | ||
| ρ01participantID | 0.45 | ||
| ICC | 0.40 | ||
| NParticipantID | 80 | ||
| NItem | 10 | ||
| Observations | 800 | ||
| Marginal R2/conditional R2 | 0.043/0.423 | ||
glmer-model results of responses to content questions.
Correct responses to content questions were coded as 1, incorrect responses as 0. Signif. codes: *** <0.001, ** <0.01.
Figure 2
5.3.2 Comprehensibility ratings
The ratings on the two comprehensibility scales showed a significant moderate positive correlation [rs(2, 398) = 0.423, p = <0.001].
The analysis (Table 4, Model M 2.1) yielded a significant main effect of LexTALE scores on sentence difficulty ratings: the higher a participant's LexTALE score, i.e., the higher their proficiency level in German, the easier they rated the sentences of the text—independent of which gender form the texts were formulated in and the attitudes toward GFL participants held (Figure 3). Note that, even though the model suggests that gender form might have a marginally significant effect on sentence difficulty ratings, we decided to not interpret this effect as significant for the following reason: after data collection, we identified a minor error in the Qualtrics stimuli presentation for two texts in the gender star condition: Klavierspieler*innen and Gitarrenspieler*innen. Specifically, one and five words, respectively, were missing at the end of each text, but no target words were affected. Due to this error, we ran the same statistical analyses without these two texts in both the star and generic masculine conditions (see supplementary materials on OSF; Zacharski et al., 2026). For all outcome variables, the results remained consistent, with one exception: the previously marginally significant effect of gender form on sentence difficulty disappeared.
Table 4
| Predictors | M 2.1 sentence difficulty ratings | M 2.2 subjective comprehensibility ratings | ||||
|---|---|---|---|---|---|---|
| Estimates | CI | p-value | Estimates | CI | p-value | |
| (Intercept) | −0.06 | −0.21–0.10 | 0.451 | −0.02 | −0.20–0.16 | 0.804 |
| Gender form | 0.11 | −0.00–0.23 | 0.051∧ | 0.04 | −0.08–0.16 | 0.525 |
| LexTALE | −0.18 | −0.30–−0.06 | 0.004** | −0.14 | −0.24–−0.05 | 0.004** |
| Attitude GFL | −0.05 | −0.18–0.08 | 0.420 | 0.01 | −0.09–0.11 | 0.855 |
| Gender form × LexTALE | 0.06 | −0.05–0.17 | 0.287 | 0.05 | −0.06–0.16 | 0.415 |
| Gender form × attitude GFL | −0.09 | −0.20–0.03 | 0.126 | −0.07 | −0.19–0.04 | 0.228 |
| LexTALE × attitude GFL | 0.02 | −0.09–0.14 | 0.683 | 0.02 | −0.07–0.11 | 0.651 |
| 15.6-7.4,-14.3498pt(Gender form × LexTALE) × attitude GFL | −0.04 | −0.14–0.07 | 0.486 | −0.00 | −0.11–0.10 | 0.948 |
| Random effects | ||||||
| σ2 | 0.30 | 0.41 | ||||
| τ00participantID:Item | 0.43 | 0.47 | ||||
| τ00participantID | 0.18 | 0.06 | ||||
| τ00item | 0.02 | 0.05 | ||||
| τ11participantID × genderform | 0.02 | |||||
| ρ01participantID | 0.71 | |||||
| ICC | 0.69 | 0.58 | ||||
| NparticipantID | 80 | 80 | ||||
| Nitem | 10 | 10 | ||||
| Observations | 2,400 | 2,400 | ||||
| Marginal R2/conditional R2 | 0.041/0.707 | 0.018/0.591 | ||||
lmer-model results of sentence difficulty ratings (Model M 2.1) and subjective text comprehensibility ratings (Model M 2.2).
Signif. codes: ** <0.01, ∧ <0.1.
Figure 3
Analogous to sentence difficulty ratings, the analysis of subjective comprehensibility ratings (Table 4, Model M 2.2) yielded a significant main effect of German proficiency: participants with higher LexTALE scores, i.e., a higher German proficiency, had an easier time understanding the texts presented—independent of gender form and attitudes toward GFL (Figure 3).
When manually checking students' justifications for their comprehensibility ratings, we found that students mainly commented on syntax, vocabulary and text structure without referring to the gender star. Only two out of 80 students mentioned the gender star as a factor impeding their reading flow when reading a text in which it was used. These limited references to the gender star as a source of difficulty are consistent with the quantitative findings, which demonstrated no significant effect of this gender-fair form on comprehension.
5.3.3 Reading times
On average, participants took M = 78.6 (SD = 31.3, range of participants' mean RT: 39.0–132) to read a text. The analysis of total text reading times (Table 5, Model M3) yielded no significant effects. In particular, no significant differences in reading times were found between texts in masculine form (M = 77.7, SD = 30.1, range: 23.5–199) and star form (M = 79.4, SD = 32.5, range: 20.9–191).
Table 5
| Predictors | M3 log (reading times) | ||
|---|---|---|---|
| Estimates | CI | p-value | |
| (Intercept) | 4.29 | 4.21–4.36 | <0.001*** |
| Gender form | 0.01 | −0.03–0.05 | 0.662 |
| LexTALE | 0.00 | −0.06–0.07 | 0.965 |
| Attitude GFL | −0.02 | −0.09–0.05 | 0.606 |
| Gender form × LexTALE | 0.00 | −0.04–0.04 | 0.879 |
| Gender form × attitude GFL | 0.01 | −0.03–0.05 | 0.525 |
| LexTALE × attitude GFL | −0.02 | −0.08–0.05 | 0.640 |
| 15.6-7.4,-14.3242pt(Gender form × LexTALE) × attitude GFL | 0.02 | −0.02–0.06 | 0.316 |
| Random effects | |||
| σ2 | 0.08 | ||
| τ00participantID | 0.08 | ||
| τ00item | 0.00 | ||
| ICC | 0.51 | ||
| NparticipantID | 80 | ||
| Nitem | 10 | ||
| Observations | 759 | ||
| Marginal R2/conditional R2 | 0.003 / 0.510 | ||
lmer-model results for the analysis of total text reading times (M3).
Signif. codes: *** <0.001.
6 Discussion
This study experimentally examined the comprehensibility of GFL for learners of German by comparing content question accuracy, subjective comprehensibility ratings, and full-text reading times between texts that use the gender star to texts that use only masculine forms. Based on previous research, we expected that learners of German would face difficulties when reading texts with the gender star form (H1), which is not in line with German orthography: in a lexical decision task, Zacharski et al. (2025) found slower reaction times for the recognition of star words among non-students compared to students, although they also observed a rapid habituation effect. They attributed these results to non-students being less familiar with the gender star. Similarly, learners of German are likely less familiar with gender star forms compared to L1 speakers, given that they acquired the language later in life and that DaF courses have been found to only marginally address the topic of GFL (e.g., ; ; Stark, 2022). This may be even more pronounced among learners whose L1 does not strongly encode grammatical gender. However, we did not find convincing support for our first hypothesis. Our results suggest that, at least for our sample of Flemish students learning German, the gender star does not significantly hinder text comprehensibility, as it did not negatively affect content question accuracy (H1a), subjective comprehensibility (H1b), sentence difficulty (H1c), or full-text reading times (H1d). This finding is particularly noteworthy given that texts in the gender star condition contained four gender star forms each, accounting for 3% to 4% of all tokens per text. However, this proportion is higher than what is typically found in journalistic texts, where gender star forms are less frequent. found that GFL forms in general take up an average of 0.95% of all tokens in journalistic texts. Hence, even in texts in which the gender star is used in a prominent manner, it did not cause major impairments of comprehensibility. Students' justifications for their comprehensibility ratings support these quantitative results, with only two out of 80 students mentioning the gender star explicitly as a factor impeding their reading flow.
In addition, we were interested in the role of two interindividual variables: students' level of proficiency in German, and their attitudes toward GFL. With regard to the former, we indeed found a main effect of proficiency on text comprehensibility measures (H2a) in the sense that students with a lower proficiency performed less well on content question accuracy and also perceived the texts as being less comprehensible. In addition, less proficient learners were also less likely to already be familiar with the gender star compared to students with a higher German proficiency as suggested by a positive correlation between proficiency and familiarity. Contrary to our expectation as formulated in H2b, however, we did not find an interaction effect of proficiency and gender form of the noun (gender star vs. masculine) on comprehensibility measures. This means that lower German proficiency negatively affected comprehensibility measures independent of the gender forms that were used in the texts, or in other words, lower German proficiency (and thus also lower familiarity with the gender star) did not result in more pronounced comprehensibility issues for texts with the gender star compared to texts with the masculine form. By contrast, higher proficiency was associated with generally improved comprehension, suggesting that advanced learners are better equipped to process a range of linguistic forms, including, for example, forms of GFL.
The second interindividual variable of interest was students' attitudes toward GFL. We chose not to formulate a hypothesis on a potential interaction effect between attitudes and gender form on comprehensibility measures, given the fact that previous research on this topic came to contradictory results (Steiger-Loerbroks and von Stockhausen, 2014; ; Zacharski et al., 2025). Similar to and Zacharski et al. (2025), we did not find an effect of attitudes toward GFL on the comprehensibility of gender star forms. Put differently, whether students had more or less positive attitudes toward GFL did not affect their comprehension of texts with the gender star vs. the masculine forms. This may be due to the fact that comprehensibility relates to cognition and should in that sense be considered as something separate from the field of opinions and emotions (Renström et al., 2022; ).
All in all, the criticism that the gender star may hinder text comprehensibility and poses a burden on learners of German by complicating an already complex language has not been confirmed in the present study. In the present sample of young Flemish learners of German, we found—consistent with previous findings for L1 speakers (; ; Zacharski et al., 2024, 2025)—that gender star words do not impair text comprehensibility. By examining this topic, our study addresses a research gap as identified by , , and Zacharski et al. (2025).
Our results have several implications for the DaF classroom. The observation that only half of the students in our sample were already familiar with the gender star, confirms previous observations that GFL is not a prominent topic in the DaF classroom (e.g., ; ; Stark, 2022). However, our study suggests that the gender star does not pose a significant burden for learners of German—at least in our sample of young Flemish students. We therefore recommend that teachers address and integrate the gender star in DaF instruction, even at early stages of language acquisition. While learners' proficiency levels did affect their overall text comprehension, there was no difference in performance between texts using the gender star and those using masculine forms. This indicates that the gender star is not more difficult for lower-proficiency learners, while higher proficiency appears to support more efficient processing across linguistic forms. We therefore believe that introducing students to the gender star early on is beneficial, as it better prepares them to communicate professionally and inclusively in German.
The study is the first experimental study aiming at understanding potential hindrances GFL may pose for language learners, thereby addressing an important research gap. A particular strength of the study lies in its assessment of comprehensibility in a natural classroom setting, which allows for a realistic reading and learning environment. However, several limitations should be acknowledged and addressed in future research. First, our sample consisted primarily of young, predominantly female students. Future studies should aim for greater diversity in terms of age, educational background, and gender identity. In particular, testing older learners of German could help disentangle the effects of familiarity with GFL from age-related processing differences, as described by Zacharski et al. (2025) for L1 speakers. Second, our findings are based primarily on explicit measures. While this approach allowed us to maintain a real-life classroom setting, prior research with L1 speakers has shown that initial processing difficulties may not be captured by such measures. For example, while found no processing difficulties in older participants using subjective ratings, Zacharski et al. (2025) identified initial difficulties using an implicit method. Although full-text reading times in our study did not indicate any processing issues related to the gender star, future research should replicate the study using eye-tracking—widely regarded as the method of choice for investigating real-time reading processes—to detect potential effects such as early processing costs that may not be visible through explicit measures. Finally, most research on GFL has focused on comprehension rather than production. Further studies involving learners could provide valuable insights into how early exposure to GFL influences learners' ability to produce inclusive language, and whether it facilitates more natural and confident use over time.
Statements
Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.
Ethics statement
The studies involving humans were approved by the Ethics Committee of the Faculty of Arts and Philosophy at Ghent University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
SD: Resources, Investigation, Writing – original draft, Writing – review & editing, Data curation, Validation, Methodology, Project administration, Conceptualization, Supervision. LZ: Methodology, Validation, Conceptualization, Writing – review & editing, Data curation, Funding acquisition, Software, Formal analysis, Writing – original draft, Resources, Project administration, Visualization. GB: Writing – review & editing, Data curation, Project administration, Supervision, Investigation, Methodology. EF: Resources, Conceptualization, Methodology, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. Lisa Zacharski's work was partly supported by the German Research Foundation (DFG) (project number: 456835372/Ferstl, FE 474/5-1). Subsequent work was conducted as part of their institutional position, with no additional external funding. For all other authors, this work was conducted within the framework of the authors' institutional positions.
Acknowledgments
We would like to thank master's students Camille Neirynck, Dagmar Dermul, and Jade Dewitte for collecting the data as part of their theses. We are also grateful to the lecturers at the participating universities and universities of applied sciences for granting access to their classes and allowing us to collect data during class time.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AckeH. (2019). Sprachwandel durch feministische Sprachkritik: Geschlechtergerechter Sprachgebrauch an den Berliner Universitäten. Z. Literaturwissenschaft Linguist.49, 303–320. doi: 10.1007/s41244-019-00135-1
2
BatesD.KlieglR.VasishthS.BaayenH. (2015a). Parsimonious Mixed Models (version 2). arXiv doi: 10.48550/arXiv.1506.04967
3
BatesD.MächlerM.BolkerB.WalkerS. (2015b). Fitting linear mixed-effects models using lme4. J. Stat. Softw.67, 1–48. doi: 10.18637/jss.v067.i01
4
BlakeC.KlimmtC. (2010). Geschlechtergerechte Formulierungen in Nachrichtentexten. Publizistik55, 289–304. doi: 10.1007/s11616-010-0093-2
5
BonvinA.BruggerL.BertheleR. (2023). Lexical measures as a proxy for bilingual language dominance?Int. Rev. Appl. Linguist. Lang. Teach.61, 257–285. doi: 10.1515/iral-2020-0093
6
BraunF.GottburgsenA.SczesnyS.StahlbergD. (1998). Können Geophysiker Frauen sein? Generische Personenbezeichnungen im Deutschen. Z. Ger. Linguist.26, 265–283. doi: 10.1515/zfgl.1998.26.3.265
7
Bundesministerium des Innern und für Heimat (2018). Zusätzliche Geschlechtsbezeichnung “divers” für Intersexuelle eingeführt. Gesetz zur Änderung der in das Geburtenregister einzutragenden Angaben. Available online at: https://www.bmi.bund.de/SharedDocs/pressemitteilungen/DE/2018/12/drittes-geschlecht.html (Accessed February 4, 2026).
8
DecockS.Van HoofS.SoensE.VerhaegenH. (2024). The comprehensibility and appreciation of non-binary pronouns in newspaper reporting. The case of hen and die in Dutch. Appl. Linguist.45, 330–347. doi: 10.1093/applin/amad028
9
ElmigerD.Schaeffer-LacroixE.TungerV. (2017). Geschlechtergerechte Sprache in Schweizer Behördentexten: Möglichkeiten und Grenzen einer mehrsprachigen Umsetzung. Osnabr. Beitr. Sprachtheor.90, 61–90. doi: 10.17192/obst.2017.90.8822
10
FerstlE.NüblingD. (2024). “Sonderzeichen als typographische Kennzeichnung geschlechtersensiblen Sprachgebrauchs: Linguistische Überlegungen und experimentelle Befunde zum Gender star,” in Orthographie in Wissenschaft und Gesellschaft: Schriftsystem – Norm – Schreibgebrauch, eds. S. Krome, M. Habermann, H. Lobin, and A. Wöllstein (Berlin: De Gruyter), 259–284. doi: 10.1515/9783111389219-014
11
FitzpatrickZ. R.JohnsonC.RottS. (2025). From society to the screen: navigating non-binary inclusion in the German-language classroom. Die Unterrichtspraxs/Teach. Ger.58, 240–249. doi: 10.1111/tger.70024
12
FoertschJ.GernsbacherM. A. (1997). In search of gender neutrality: Is singular they a cognitively efficient substitute for generic he?Psychol. Sci.8, 106–111. doi: 10.1111/j.1467-9280.1997.tb00691.x
13
FriedrichM. (2017). Textverständlichkeit und ihre Messung: Entwicklung und Erprobung eines Fragebogens zur Textverständlichkeit. Münster: Waxmann.
14
FriedrichM. C. G.DrößlerV.OberlehbergN.HeiseE. (2021). The influence of the gender asterisk (“Gendersternchen”) on comprehensibility and interest. Front. Psychol.12:760062. doi: 10.3389/fpsyg.2021.760062
15
FriedrichM. C. G.HeiseE. (2019). Does the use of gender-fair language influence the comprehensibility of texts?Swiss J. Psychol.78, 51–60. doi: 10.1024/1421-0185/a000223
16
GabrielU.GygaxP. M.KuhnE. A. (2018). Neutralising linguistic sexism: Promising but cumbersome?Group Process. Intergr. Relat.21, 844–858. doi: 10.1177/1368430218771742
17
GlimS.KörnerA.HärtlH.RummerR. (2023). Early ERP indices of gender-biased processing elicited by generic masculine role nouns and the feminine–masculine pair form. Brain Lang.242:105290. doi: 10.1016/j.bandl.2023.105290
18
GreenP.MacLeodC. J. (2016). SIMR: an R package for power analysis of generalized linear mixed models by simulation. Methods Ecol. Evol.7, 493–498. doi: 10.1111/2041-210X.12504
19
GygaxP.ElmigerD.ZuffereyS.GarnhamA.SczesnyS.Von StockhausenL.et al. (2019). A Language index of grammatical gender dimensions to study the impact of grammatical gender on the way we perceive women and men. Front. Psychol.10:1604. doi: 10.3389/fpsyg.2019.01604
20
GygaxP.SatoS.ÖttlA.GabrielU. (2021). The masculine form in grammatically gendered languages and its multiple interpretations: a challenge for our cognitive system. Lang. Sci.83:101328. doi: 10.1016/j.langsci.2020.101328
21
IrmenL. (2007). What's in a (role) name? Formal and conceptual aspects of comprehending personal nouns. J. Psycholinguist. Res.36, 431–456. doi: 10.1007/s10936-007-9053-z
22
KeithN.HartwigK.RichterT. (2022). Ladies first or ladies last: do masculine generics evoke a reduced and later retrieval of female exemplars?Collabra Psychol.8:32964. doi: 10.1525/collabra.32964
23
KerschenK.ScottC.E.WalterD. (2025). Between and beyond er|sie: trans and non-binary identities in the L2 German classroom. Die Unterrichtspraxs/Teach. Ger. 58 169–174. doi: 10.1111/tger.70038
24
KintschW. (1988). The role of knowledge in discourse comprehension: a construction-integration model. Psychol. Rev.95, 163–182. doi: 10.1037/0033-295X.95.2.163
25
KleinS. S. (1988). Sex education and gender equity. Educ. Leadersh.45:69.
26
KörnerA.AbrahamB.RummerR.StrackF. (2022). Gender representations elicited by the gender star form. J. Lang. Soc. Psychol.41, 553–571. doi: 10.1177/0261927X221080181
27
KotthoffH. (2020). Gender-Sternchen, Binnen-I oder generisches Maskulinum, … (Akademische) Textstile der Personenreferenz als Registrierungen?Linguist. Online103, 105–127. doi: 10.13092/lo.103.7181
28
KotthoffH.NüblingD. (2024). Genderlinguistik: Eine Einführung in Sprache, Gespräch und Geschlecht (2., überarbeitete und erweiterte Auflage). Tübingen: Narr Francke Attempto. doi: 10.24053/9783381105922
29
KromeS. (2022). Gendern in der Schule: Zwischen Sprachwandel und orthografischer Norm. Mitt. Dtsch. Germanistenverb.69, 86–110. doi: 10.14220/mdge.2022.69.1.86
30
KromeS. (2020). Zwischen gesellschaftlichem Diskurs und Rechtschreibnormierung: Ge-schlechtergerechte Schreibung als Herausforderung für gelungene Textrealisation. Sprachdienst64, 31–45.
31
KurzP.De MulderH. (2023). A star is born? The German gender star and its effects on mental representation. Psychol. Lang. Commun.27, 384–404. doi: 10.58734/plc-2023-0018
32
LemhöferK.BroersmaM. (2012). Introducing LexTALE: a quick and valid lexical test for advanced learners of English. Behav. Res. Methods44, 325–343. doi: 10.3758/s13428-011-0146-0
33
LinkS. (2023). “Gendergerechte Sprache: Präsenz in den deutschen Medien und Integration in den DaF-Unterricht,” in Prospettive discorsive e di educazione linguistica internazionale, ed. N. Leal Rivas (Napoli: Federico II University Press), 45–63.
34
LipskyA. (2023). Gendersensible Grammatik: Vorschläge zur Vermittlung von Genus und Personenbezeichnungen. Inf. Deutsch. Fremdspr.50, 543–560. doi: 10.1515/infodaf-2023-0085
35
MerkelE. (2011). Geschlechtergerechte Sprache in den universitären Gleichstellungskonzepten – eine vergleichende Analyse. J. Netzw. Frauen Geschlechterforsch.28, 36–41.
36
MeuleneersP. (2022). “Reduktionen in der medialen Debatte um die Gendersprache.” in GfdS Lecture (online) [Conference presentation] 28 November 2022 (Wiesbaden).
37
MeuleneersP. (2024). “On the “invention” of the Gendersprache in German media discourse,” in Public Attitudes towards Gender-Inclusive Language, ed. F. Pfalzgraf (Berlin: de Gruyter Mouton), 159–182. doi: 10.1515/9783111202280-007
38
MiserskyJ.GygaxP. M.CanalP.GabrielU.GarnhamA.BraunF.et al. (2014). Norms on the gender perception of role nouns in Czech, English, French, German, Italian, Norwegian, and Slovak. Behav. Res. Methods46, 841–871. doi: 10.3758/s13428-013-0409-z
39
Müller-SpitzerC.OchsS. (2024). Shifting social norms as a driving force for linguistic change: Struggles about language and gender in the German Bundestag. arXiv doi: 10.48550/arXiv.2402.03887
40
Müller-SpitzerC.OchsS.KoplenigA.RüdigerJ. O.WolferS. (2024). Less than one percent of words would be affected by gender-inclusive language in German press texts. Humanit. Soc. Sci. Commun.11:1343. doi: 10.1057/s41599-024-03769-w
41
Müller-SpitzerC. (2022). Gendergerechter Sprachgebrauch. Ein komplexes Feld für die Aushandlung von ?richtig‘ und ?gut‘. Deutschunterricht75, 39–49.
42
PabstL. M.KollmayerM. (2023). How to make a difference: the impact of gender-fair language on text comprehensibility amongst adults with and without an academic background. Front. Psychol.14:1234860. doi: 10.3389/fpsyg.2023.1234860
43
PaulA. M. (2025). Gender-inclusive language instruction in the German classroom. Die Unterrichtspraxs/Teach. Ger.58, 228–239. doi: 10.1111/tger.70034
44
PeuschelK. (2022). Keine größere Hürde als...gendergerechte Sprache im Deutschen aus der Perspektive des Lehrens und Lernens. Polit. Zeitgesch.72, 49–50.
45
PeuschelK.SchmidtL. (2022). “Gendergerechte Sprache in Deutsch als Fremd- und Zweitsprache: Einstellungen von Studierenden zu ihrem Gebrauch in Universität und Unterrichtspraxis, ” in Gender_Vielfalt_Sexualität(en) im Fach Deutsch als Fremd- und Zweitsprache, eds. A. Freese, and O. N. Völkel (Munich: Iudicium), 53–72. Available online at: https://opus.bibliothek.uni-augsburg.de/opus4/frontdoor/deliver/index/docId/83900/file/83900.pdf (Accessed May 14, 2026).
46
PöschkoH.PrielerV. (2018). Zur Verständlichkeit und Lesbarkeit von geschlechtergerecht formulierten Schulbuchtexten. Z. Bild. Forsch.8, 5–18. doi: 10.1007/s35834-017-0195-2
47
Posit Team (2023). RStudio: Integrated Development Environment for R. Posit Software, PBC. Available online at: http://www.posit.co/ (Accessed February 4, 2026).
48
PreseauL. (2025). A comparative analysis of gender-just language in first-year German textbooks. Die Unterrichtspraxs/Teach. Ger.58, 301–310. doi: 10.1111/tger.70030
49
PuschL. F. (1984). Das Deutsche als Männersprache.Frankfurt am Main: Suhrkamp.
50
Rat für deutsche Rechtschreibung (2021). Geschlechtergerechte Schreibung: Empfehlungen vom 26.03.2021. Available online at: https://www.rechtschreibrat.com/DOX/rfdr_PM_2021-03-26_Geschlechtergerechte_Schreibung.pdf (Accessed February 4, 2026).
51
RenströmE. A.LindqvistA.SendénM. G. (2022). The multiple meanings of the gender-inclusive pronoun hen: Predicting attitudes and use. Eur. J. Soc. Psychol.52, 71–90. doi: 10.1002/ejsp.2816
52
SatoS.GygaxP. M.GabrielU. (2016). Gauging the impact of gender grammaticization in different languages: application of a linguistic-visual paradigm. Front. Psychol.7:140. doi: 10.3389/fpsyg.2016.00140
53
SchneiderJ. G. (2022). “Gendern in institutionellen Leitfäden. Im Spannungsfeld von Indexikalität und grammatischen Erfordernissen,” in Ratgeben in der spätmodernen Gesellschaft. Ansätze einer linguistischen Ratgeberforschung, eds. M. Hennig, and R. Niemann (Tübingen: Stauffenburg), 233–261.
54
SchunackS.BinanzerA. (2022). Revisiting gender-fair language and stereotypes – a comparison of word pairs, capital I forms and the asterisk. Z. Sprachwiss.41, 309–337. doi: 10.1515/zfs-2022-2008
55
SpeyerL. G.SchleefE. (2019). Processing ‘gender-neutral'pronouns: a self-paced reading study of learners of English. Appl. Linguist.40, 793–815. doi: 10.1093/applin/amy022
56
StahlbergD.SczesnyS. (2001). Effekte des generischen Maskulinums und alternativer Sprachformen auf den gedanklichen Einbezug von Frauen. Psychol. Rundsch.52, 131–140. doi: 10.1026//0033-3042.52.3.131
57
StahlbergD.SczesnyS.BraunF. (2001). Name your favorite musician: effects of masculine generics and of their alternatives in German. J. Lang. Soc. Psychol.20, 464–469. doi: 10.1177/0261927X01020004004
58
StarkK. S. (2022). Bewertung und Vermittlung gendergerechter Sprache durch DaZ/DaF-Lehrkräfte: evaluation and teaching of gender-sensitive language by GSL/GFL teachers. Inf. Deutsch. Fremdspr.49, 557–574. doi: 10.1515/infodaf-2022-0055
59
SteckerC.MüllerJ.BlätteA.LeonhardtC. (2021). The evolution of gender-inclusive language. Evidence from the German Bundestag, 1949-2021. OSF Preprints doi: 10.31219/osf.io/fcsmz
60
Steiger-LoerbroksV.von StockhausenL. (2014). Mental representations of gender-fair nouns in German legal language: an eye-movement and questionnaire-based study. Linguist. Ber.237, 57–80. doi: 10.46771/2366077500237_4
61
SteinhauerA.DiewaldG. (2017). Richtig gendern: Wie Sie angemessen und verständlich schreiben (1st ed). Berlin: Duden Verlag.
62
StummerN. (2024). Gendersensible Sprache im Fremdsprachenunterricht Deutsch: Aktuelle Befunde aus Lehrmaterialien. Wien. Linguist. Gaz.96, 1–28.
63
ToonenJ. (2022). Geschlechtsübergreifende Personenbezeichnungen in wissenschaftlichen Zeitschriftenartikeln: Ein deutsch-niederl?ndischer Vergleich. Linguist. Online113, 153–188. doi: 10.13092/lo.113.8351
64
Trömel-PlötzS. (1997). “From male debates to women's conversations: a journey in feminist linguistics,” in Kommunikation von Geschlecht, eds. F. Braun, and U. Paero (Herbolzheim: Centaurus), 336–349.
65
TruanN. (2023). “In Deutschland wird es weniger diskutiert”: Sprachideologien über geschlechtergerechte Sprache aus der Sicht von L2-Sprechenden des Deutschen. Linguist. Ber.36, 237–257.
66
Van HoofS.DecockS. (2026). “Gender-inclusive language in Dutch,” in Gender-Inclusive Language: Findings from 14 Languages and Open Research Questions, ed. F. Pfalzgraf (Berlin: de Gruyter Brill), 41–67. doi: 10.1515/9783111701608-003
67
VerveckenD.HannoverB.WolterI. (2013). Changing (S)expectations: how gender fair job descriptions impact children's perceptions and interest regarding traditionally male occupations. J. Vocat. Behav.82, 208–220. doi: 10.1016/j.jvb.2013.01.008
68
WaldendorfA. (2024). Words of change: the increase of gender-inclusive language in German media. Eur. Sociol. Rev.40, 357–374. doi: 10.1093/esr/jcad044
69
Westdeutscher Rundfunk and infratest dimap (2022). Gendergerechte Sprache [Public survey]. Available online at: https://www.infratest-dimap.de/umfragen-analysen/bundesweit/umfragen/aktuell/gendergerechte-sprache/ (Accessed February 4, 2026).
70
WootenA. (2025). To gender or not to gender?: Strategies for teaching gender-inclusive language in the L2 German classroom. Die Unterrichtspraxs/Teach. Ger.58, 221–227. doi: 10.1111/tger.70021
71
YouGov (2023). Finden Sie geschlechtergerechte Sprache, sogenanntes Gendern (z.B. Lehrer*in / LehrerIn), wichtig oder unwichtig? [Public survey]. Available online at: https://yougov.de/topics/society/survey-results/daily/2023/03/07/5cbaa/2 (Accessed February 4, 2026).
72
ZacharskiL. (2024). “Using pair forms, criticizing the gender star – attitudes towards binary and non-binary gender-inclusive language in German,” in Public Attitudes towards Gender-Inclusive Language, ed. F. Pfalzgraf (Berlin: De Gruyter Mouton), 209–242. doi: 10.1515/9783111202280-009
73
ZacharskiL.FerstlE. C. (2023). Gendered representations of person referents activated by the nonbinary gender star in German: a word-picture matching task. Discourse Process.60, 294–319. doi: 10.1080/0163853X.2023.2199531
74
ZacharskiL.KruppaA.FerstlE. C. (2025). The readability of the non-binary gender star in German: evidence from a lexical decision gask. Soc. Psychol. Bull.20:e13719. doi: 10.32872/spb.13719
75
ZacharskiL.SudermannT.KappS.FerstlE. C. (2024). “Does the use of non-orthographic gender-fair forms impair the readability of texts? Insights from an eye-tracking study [Conference presentation],” in 34th Annual Meeting of the Society of Text and Discourse 2024 (STandD 2024). July 17–19, 2024 (Chicago, IL).
76
ZacharskiL.DecockS.BooneG. (2026). Comprehensibility of Gender-Fair Language Among Learners of German [Online Repository]. Available online at: https://osf.io/g6bsv?view_only=4ad6ac363bbd429fb93641220c264f96 [view-only link for peer review].
Summary
Keywords
comprehensibility, gender star, gender-fair language, gender-inclusive language, L2 German, LX German, nonbinary gender forms
Citation
Decock S, Zacharski L, Boone G and Ferstl EC (2026) Comprehensibility of gender-fair language among foreign language learners of German: an experimental study. Front. Lang. Sci. 5:1806497. doi: 10.3389/flang.2026.1806497
Received
07 February 2026
Revised
27 April 2026
Accepted
30 April 2026
Published
28 May 2026
Volume
5 - 2026
Edited by
Joanna Chojnicka, University of Groningen, Netherlands
Reviewed by
Anestis Karastergiou, National and Kapodistrian University of Athens, Greece
Morana Lukač, University of Groningen, Netherlands
Updates
Copyright
© 2026 Decock, Zacharski, Boone and Ferstl.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Lisa Zacharski, lisa.zacharski@uni-bremen.de; Sofie Decock, sofie.decock@ugent.be
† These authors share first authorship
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.