An Investigation on the Effects of Ambiguity, Gender Orientation, and Domain Relatedness of Design Projects on Student Performance

1 Gül E. Kremer, gkremer@iastate.edu Ac ce pt ed M an us cr ip t N ot C op ye di te d D ow naded rom htt://asm edigitallection.asm erg/m echanicaldesign/arti/10.1115/1.4045300/6434395/m d-18705.pdf by ow a State U niersity user on 11 Feruary 2020 Journal of Mechanical Design 2 Students in design courses work on projects that are influenced by ambiguity, gender orientation, and domain relatedness. This study investigates the impacts of these factors on student self-efficacy in order to increase retention in engineering disciplines. From a comprehensive literature review and feedback from engineering experts, an instrument is developed to assess student perceptions on tolerance to ambiguity (STA), project gender orientation (PGO) and project domain relatedness (PDR). Statistical analyses are conducted to examine the influence of STA, PGO, and PDR on student self-efficacy and collective-efficacy. Results indicate that an increase in the gender orientation of the project decreases student self-efficacy. Furthermore, gender bias of the design project diminishes student tolerance to deal with ambiguous situations. Therefore, instructors should consider choosing more gender-neutral projects or make appropriate adjustments in project descriptions to minimize gender bias.


INTRODUCTION
There is now wide understanding that the "sage on the stage" style of instruction has limitations in appropriately preparing students for the modern workplace. In the "sage on the stage" or transmittal form of teaching, King [1] assumes that the student brain can be likened to a container into which the professor pours in knowledge. In contrast, the constructivist theory of learning sees the instructor as the "guide on the side" facilitating student interaction and experience with the material, enabling knowledge construction and meaning to take place [2]. Project-based design learning environment provides this facilitated learning setting [3], and thus it is being used throughout the engineering curriculum to provide design experiences to students [4]. In many universities, engineering design courses feature project-based learning starting in the first year.
Several authors note that the structure of the curriculum can promote interdisciplinary collaborations [5][6][7]. Project-based design courses can bring students from various disciplines into a team. At the first year, another advantage of a projectbased design course setting is it provides a platform for students to experience engineering fundamentals together even if their eventual disciplinary choices (i.e., chemical engineering vs. industrial engineering) are different. Moreover, design projects can enhance students' motivation by increasing the enthusiasm of first year students through engagement with authentic industrial clients and problems originating from real-world applications. Design projects coming from industry provide an opportunity to see the real applications of their design solutions, and how the theory and practice come together.
There are numerous studies in the literature that attest to the benefits of project-based learning within design courses such as increasing the motivation, team working and communication skills [8], retention [9], and ability to implement design applications [4,10]. Despite these advantages, some of which are provided above, if not designed well such project-based design experiences might have negative impacts on student learning and motivation. For example, Jones at el. [11] showed that at the end of the first year, engineering students' expectancy and value related motivation constructs such as self-efficacy, success expectation, beliefs on importance of engineering, and usefulness of the engineering field decrease. An inappropriate design project may be the reason for the decrease in self-efficacy and students' belief on the importance of the engineering field, since it is the only course where most freshmen orientation" and "major relatedness" of the design project may increase this attrition rate from engineering fields to other disciplines due to these factors being seen as barriers to student success in design projects. Although their study provides preliminary results based on student views and comments, a comprehensive study is needed to clarify the effects of these potential barriers on student success in a design course. In this paper, we focus on gender orientation of design projects, relatedness of the project domain with the discipline that students want to pursue, and student preparedness to tackle open-ended and potentially ambiguous design projects as barriers to their success in design projects.
A design project may be skewed toward a gender due to its masculine or feminine tasks or subject matter, unintentionally disenfranchising one of the gender groups in the design learning setting [15]. For example, if a design project mainly focusses on male related interests such as designing the engine of a vehicle, it may decrease the motivation of female students and discourage them from their pursuit of engineering as a career. Confirming this for males, they were shown to be unwilling to improve their competencies if they thought the project was related to feminine tasks [16]. Although it is important to understand the impact of gender orientation of the design project on student learning and performance, extant research on this topic is limited in many ways.
Major or domain relatedness of the design project is another potential barrier, which may decrease student motivation in design courses. When the project relates to a specific domain, students who want to pursue a career in that engineering specialization may feel more engaged while others may feel less motivated and lose interest. Richter and Paretti [7] showed that major bias of the project negatively affects student performance in interdisciplinary teams, specifically for those who do not want to get a degree in that field. Design projects should have multidisciplinary appeal so that both teamwork skills and the commitment of engineering students can be improved.
In the first year, engineering design projects are conducted by students who may lack sufficient domain specific knowledge. Moreover, when the projects originate from real-world applications, they are usually less structured and abstract and thus require open-ended complex problem solving skills from students. The ambiguity due to the lack of adequate domain knowledge or this unstructured and abstract form of the design project may cause loss of motivation, resulting in a less than ideal learning setting for many engineering students. Student interests and the characteristics of the academic disciplines may also affect their tolerance to cope with barriers rooted in the design project description.
Holland's theory may be used to propose a link between student interests with the characteristics of academic disciplines. Holland's theory focuses on assessment of individuals, their environments (i.e., academic major) and the interaction between the two. According to this theory, an individual's selection of academic major is an expression of their interests and preferences. Per this theory, most people can be classified into one or more of the six theoretical types: realistic, investigative, artistic, social, enterprising, and conventional [17,18]. For example, electrical and mechanical engineering are classified as realistic majors, whereas industrial engineering is enterprising [18]. As per Holland's classification student characteristics across groups vary, and thus there may be differences among majors in coping with ambiguity. For instance, since electrical and mechanical engineering students are likely to be realistic (if their major choice is congruent with their interests), their tolerance to ambiguous situations may not be as high as their peers' pursuing civil or chemical engineering who are classified as investigative. Moreover, gender specific characteristics and preferences may influence individual tolerance to ambiguity.
In this study, we investigate the following research questions: (RQ1) Which combination of predictors, STA, PGO, PDR and Holland's classification best predicts student self-efficacy in terms of expected grades? (RQ2) Which combination of the predictors STA, PGO, PDR, gender, student confidence in expected grades, and student expected grades best predicts student collective efficacy (SCE)? (RQ3) Does tolerance to ambiguity of different major groups classified by Holland and gender vary by PGO?
The goal of this work is to explore the effects of three barriers, "project ambiguity," "project gender orientation," and "project domain relatedness" on student success in engineering design projects. The results of this study will support educators in managing these potential barriers to self-efficacy and thus retention.
What follows is a summary of our review on ambiguity, major relatedness and gender bias as potential barriers to student self-efficacy and collective-efficacy. We then explain our research methodology and unfold the results organized as responses to the research questions posed. Last, we discuss the study results and limitations of the study.

REVIEW OF LITERATURE ON THE THREE BARRIERS
Our review included engineering design and engineering research manuscripts related to project ambiguity, project domain relatedness, and project gender orientation, and focused on them as potential barriers to student success in engineering design courses. We also highlighted relevant threads on student self-and collective efficacy.

Project Ambiguity
Although traditional problems (text-book problems) common in most engineering courses prepare students for different engineering applications, they are not sufficient to increase open-ended problem solving skills of students [19,20]. Therefore, many engineering programs include industry-sponsored design projects in their design courses to increase student creative thinking and open-ended problem solving skills. However, these industry-originated problems may be ambiguous for students due to lack of necessary information and a multitude of directions to pursue for solutions.
The term ambiguity refers to perceived insufficiency of information and is used to describe decisions for which the odds of an uncertain event are not precisely known [21]. Dringenberg [22] defined ambiguity as "uncertainty [that] exists about which concepts, rules, and principles are necessary for the solution or how they are organized." The ambiguity level of industry-sponsored projects may be high for first year students, since they lack specific domain knowledge, tools and jargon. On the other hand, since these students are not bound by theoretical knowledge restrictions, they may be more creative in open-ended problem solving. Hullsiek [23] revealed that a student's tolerance to ambiguity is moderated by the relationship between situation ambiguity (i.e., project ambiguity) and student creativity. Focusing on the relation of uncertainty to learning, Tauritz [24] reported that while too little or too much uncertainty blocks learning, some level of uncertainty motivates learning. Zheng et al. [25] showed how the uncertainty of the design projects could be reduced by concept selection tools in order to aid the student decision making process. Kazerounian [26] discussed that lack of ambiguity limits the creativity of the engineering students.
However, there are many studies in the literature reporting ambiguity as a problematic issue with regards to open-ended problem solving [7,19,20,22,27].
According to Kahn and Sarin [21], individuals can be categorized as ambiguity averse, ambiguity seeking, or ambiguity indifferent based on their tolerance to ambiguous situations. Mohammed et al. [28] conducted a study to see the effects of student tolerance to ambiguity on self-efficacy, collective efficacy, satisfaction with the group and conflict resolution. They showed that individuals with a higher tolerance to ambiguity acquire higher levels of collective efficacy, group satisfaction, and conflict resolution. Moreover, ambiguity-seeking students feel more motivated to work on open-ended projects rather than more straightforward projects. On the other hand, they have less self-efficacy, which was measured by students' expected grades, compared to their ambiguity averse counterparts.
Dringenberg and Wertz [29] developed an instrument to assess how ambiguity and recognizing the value of multiple perspectives affect student acceptance or resistance to cope with open-ended design problems. The goal of the instrument is to track the development of students dealing with ambiguity and recognizing the value of multiple perspectives. Their instrument allows educators to make interventions for ambiguity adverse students to increase their open-ended problem-solving skills. These authors acknowledge though, that their study presents preliminary results and more data is needed to validate their instrument.

Gender Orientation of the Design Project
When students think that the project is skewed towards one gender, they may lose interest [30] or even change their discipline. Okudan et al. [14] reported that student evaluative ratings of the design course were low when students were conducting a military related project. However, the ratings were higher for more gender-neutral projects such as supply chain assembly and a human-powered folding trailer. The reason was articulated as a potential gender bias to the military project. Additionally, Okudan et al. [15] argued when the tasks/topics students perform during a design course are perceived as gender oriented, students from the other gender lose motivation and interest to work on the project. The thing of focus and setting, including "product/object," "experience," "institution," "action," "background knowledge" and "gender composition", constitute factors that may cause the design project to be perceived as gender oriented.
Moskal [31] and Moskal et al. [32] show female students may have less selfconfidence in their first year of engineering; however, well-supported females with sound instructional methodologies may recognize their ability and be able to perform better. Amelink and Meszaros [33] demonstrate the positive experiences female students have during their education encourage them to attend and pursue engineering careers.
Despite the increase in the number of females in science and engineering, females are still underrepresented [34]. There are studies in literature to understand reasons why the ratio of females is low in the engineering field [35,36]. A wellorganized engineering design course (disinfected from biasing factors) may support to strengthen female student beliefs on their ability for success in engineering, and increase the number of females who pursue engineering as a career.

Major/Domain Relatedness of the Project
During the first year of engineering education, even if students are not admitted to a specific engineering major, they may have an interest in a specific engineering discipline.
When students perceive that the design project is not related to that discipline, their motivation in the project and thus self-efficacy may decrease. Support for this exists in prior literature. For example, Richter and Paretti [7] discussed "relatedness" as a potential barrier to student success in interdisciplinary teamwork. The term relatedness refers to a connection between a student's major of interest and the project domain.
Students are more successful when they are capable of identifying connections between their own disciplines and the project domain. Conversely, students lose motivation when they are not capable of finding similarities between their intended major and the project domain.
A project's relevance to a student's interest in a major could not be considered without potential gender biases. There are differences across engineering majors for perceived gender orientations. For instance, female students prefer to pursue careers in genetic and bioengineering, chemical, environmental, and industrial engineering. On the other hand, male students prefer to pursue careers in electronic, mechanical and civil engineering [37,38]. A project related to chemical engineering may be ideal for most of the female students on the project team, whereas it may be less than ideal for most male students since there might not be a connection between their discipline of choice and the project domain.

Self-efficacy and Collective-efficacy
Self-efficacy, introduced by Bandura, is an individual's belief on their capacity to perform a given task [39]. For teams, it is more appropriate to consider collectiveefficacy that shows a team's belief to perform as a whole [40]. For most design project settings, because students work as a team to solve industry-originated problems, collective-efficacy is as important as self-efficacy. A preliminary study [14] showed the negative effects of project ambiguity, project gender orientation and project domain relatedness on student success; and asserted that such factors may decrease student self-and collective-efficacy. However, the study featured literature review-based hypotheses along with preliminary observations from a very limited sample. A follow on study [28] discussed the effect of student tolerance to ambiguity on their self-and collective-efficacy. Extant works also connected gender to student self-efficacy [12,15,41]. However, by and large, what exists is limited in revealing complex and potentially interdependent connections between ambiguity, gender orientation, and domain relatedness factors and self-and collective-efficacy in the context of engineering design projects. Our research intends to add to the state-of-the art at this intersection.

METHODOLOGY
The purpose of this study is to collect data from first year engineering students and analyze it to understand their perceptions on how project ambiguity, domain relatedness, and gender orientation affect their self-and collective-efficacy. To achieve this goal, we first developed and validated a survey instrument. Subsequent to the data collection, binary logistic regression, linear regression and ANCOVA analyses were conducted as we sought responses to the research questions.

A c c e p t e d M a n u s c r i p t N o t C o p y e d i t e d
The survey included 64 Likert-type questions that were organized in five subscales. The On the Likert scale, a 1 indicated "strongly disagree" and a 5 indicated "strongly agree." Some questions on the survey used reverse wording to get a stronger and more valid measure. Accordingly, these items are reverse coded for the analysis. Low points on these scales (i.e., student tolerance to ambiguity, collective efficacy, student perception on project gender orientation, project's relevance to the chosen/anticipated major and project ambiguity level) show low STA, SCE, student perceptions on PGO and PDR.

Data Sources and Analysis
The Student expected grades are used as the indicator of their self-efficacy [44].
On the survey, students indicated their expected grades and confidence in achieving these expected grades. Sixty students (46.15%) indicated that they were expecting an "A," 44 (33.85%) were expecting an "A-," while 25 (19.23%) of them were expecting a "B", "B+" or "B-." None of the students expected a grade of "C+" or "C," but one student was expecting to earn a "C-." Students were confident they would achieve their expected grades; 57 (43.85%) of them were 90% confident in receiving their expected grade. We also collected data for their chosen major or anticipated major.
While there were 24 different majors chosen, six of them were preferred to a higher degree: mechanical engineering, aerospace engineering, biomedical engineering, chemical engineering, industrial engineering and civil engineering. The majority of the students (22.31%) wanted to study mechanical engineering, followed by aerospace engineering (13.85%). Close behind were biomedical and chemical engineering with 13% and 12.31%. Last were industrial engineering and civil engineering with 8.46% each.
Other majors (16.92%) were selected by less than five students and hence by a small percentage of the study population. The percentage of students who had not decided their major or answered this question was 4.61%.
After data collection, participant responses were categorized based on Holland's type classification. In this categorization, we assume students' anticipated major choice represents their dominant Holland type. Thus, students who chose mechanical and electrical engineering disciplines as their major were coded as realistic (group 1); students who chose aerospace, chemical, biomedical and civil engineering were coded as investigative (group 2); students who chose industrial, computer and communications engineering were coded as enterprising (group 3); lastly, students who chose architectural engineering were coded as artistic (group 4).

Validity analysis
Three different validity analyses (i.e., content, face and construct validity) were performed to verify if the developed survey instrument was measuring what it was intended to. The objective of these validation processes was to specify the clarity, accuracy and the relevancy of the content [41]. The content and face validity sought to uncover if the instrument covers the considered subjects sufficiently and appropriately.
In order to accomplish this, the views and feedback of engineering education domain experts from Penn State and Iowa State were collected to critically review and revise the instrument.
For the construct validity, we performed confirmatory factor analysis (CFA) instead of exploratory factor analysis (EFA) since we adopted scales from the extant  Table 1 for female and male students, separately.
A bivariate correlation matrix for the survey variables are listed in Table 2. Based on the results observed in Table 2, there are significant positive relationships between student confidence level for achieving their expected grades and STA and SCE. In addition to these relationships, there is a significant negative relationship between PGO and STA. There is also a negative relationship between PDR and PAL. STA was not related to SCE, PDR or PAL as assessed by the correlation analysis. Student confidence levels for achieving their expected grade was not significantly related to PGO, PDR or PAL. Lastly, there was no statistically significant relationship between PGO, PDR and PAL.
Collected data was used to address the following research questions: ( Table 3. According to the Hosmer-Lemeshow test, goodness of fit measure is calculated as χ 2 (8) = 11.199, p-value= 0.19 > α =0.05.
This indicates a good model fit.
Results from the logistic regression model shows that PGO negatively affects student expected grade (odds ratio=0.60; p-value<0.05), which means the increase of project gender orientation decreases the probability of expecting to receive a higher grade. One level increase on the project gender orientation decreases the odds of students expecting to receive a higher grade by 60%. In other words, when students feel that the project is skewed to one gender, the probability of expectation to receive a high grade decreases. However, STA, PDR, and Holland's classification do not have a statistically significant effect on students' expected grades (all p-value>0.05).
RQ2: Which combination of the predictors STA, PGO, PDR, gender, student confidence of expected grades, and student expected grades best predicts SCE? A linear regression model was built to examine the significant independent variables (STA, PGO, PDR, expected grade, gender and student confidence of expected grades) to predict SCE in Table 4. An independent variable, student confidence on expected grade is statistically significant to predict SCE (p-value=0.03). Other independent variables PGO, STA, PDR, expected grade and gender are not statistically significant to predict SCE.
RQ3: Does tolerance to ambiguity of different occupation and gender groups vary by PGO? Before observing the potential effect of PGO, we examined whether there was a group difference in STA among Holland's classification groups or between genders. Thus, a two-way ANOVA was performed. According to the results reported in

DISCUSSION AND CONCLUSION
Our goal in this study is to assist educators in selecting and implementing design projects in order to increase student performance and motivation. Since the effects of STA, PGO and PDR are known, instructors should pay attention to selection of design projects to avoid negative impacts of gender bias or make adjustments to project description to mitigate the potential negative effects. Three potentially problematic issues are considered as barriers to effective design courses in the literature [14,28,29], and an extensive study examining the effects of these issues on student self-and collective efficacy in an engineering design project context is reported herein. In the third statistical analysis, the ANCOVA model shows that PGO significantly changes STA. Additionally, there is a negative correlation between PGO and STA, which means that when students feel the project is skewed to one gender their tolerance to ambiguity decreases. As a consequence, all three analyses show that an increase in project gender orientation negatively influences student self-efficacy and collectiveefficacy. Moreover, it diminishes student tolerance to cope with ambiguous situations.
This study contributes to engineering design research by demonstrating how ambiguity, gender orientation and domain relatedness impact a design project. These findings articulate how specific factors effect student performance in a design course in terms of self-efficacy and collective efficacy. No research is immune to some limitations; some of these limitations for the shared work are: (i) A comparison analysis between clearly defined design projects and more abstract ones may be performed to investigate how self-efficacy and collective-efficacy of the students change. Since there was only one design project in the term, no comparison was done in this study.
(ii) The survey can be conducted in a timely fashion so that student selfefficacy can be compared at the beginning and at the end of the design course. This approach gives an opportunity to investigate how a student's belief and abilities change throughout a specific design course.
(iii) In the regression model, PGO has a marginal significant negative effect (p-value=0.09) on SCE. Therefore, increasing the number of the participants may give an opportunity to better understand the impacts of PGO on SCE.
(iv) An individual's dominant Holland type is determined based on their anticipated major. However, we readily acknowledge that any person will have characteristics corresponding to more than one type. Further investigations will be conducted to capture and study different interest types of the individuals.  Table Captions List   Table 1 Descriptive statistics for all survey scales by gender Table 2 Correlation among variables in the survey Table 3 Logistic regression analysis results Table 4 Summary of linear regression analysis for variables predicting SCE Table 5 Summary of ANOVA model for investigating the effects of gender and major on STA    5. This industrial sponsored project will positively impact the way I evaluate the course at the end of the semester when the SRTE (Student Rating of Teaching Effectiveness) is conducted. 6. The industrial sponsored project this semester has positively impacted me to stay with my major/anticipated major. 7. This industrial sponsored project will positively impact the way I evaluate the course instructor at the end of the semester when the SRTE (Student Rating of Teaching Effectiveness) is conducted. 8. I found certain aspects of the design projects to be related to my chosen discipline Project Gender Orientation (PGO) 1. The project my group worked on was associated with a masculine or feminine product or object (e.g., guns, rockets, explosives make me think of males). 2. The project my group worked on was associated with a masculine or feminine experience (e.g., cooking makes me think of females). 3. The project my group worked on was associated with a masculine or feminine institution (e.g., the military makes me think of males). 4. The project my group worked on was associated with a masculine or feminine action related (e.g., teaching makes me think of females). 5. The project my group worked on was associated with a masculine or feminine interest related (e.g., war affects everyone, but men tend to be more interested). 6. The project my group worked on was associated with a masculine or feminine idea generation (e.g., the ideas were mostly contributed by the males). 7. The project my group worked on was associated with a masculine or feminine background knowledge (e.g., females know how to socially work in a group). 8. The project my group worked on was associated with a masculine or feminine composition (e.g., the group was all male).

Appendix B
Brief explanations on the EDSGN 100 course, its, learning goals, and the description of the design project are provided below. Course relevant information was provided by Xinli Wu.
Design project description is minimally modified from a version prepared and used by Engineering Design Program faculty at Penn State.

Course Overview
EDSGN 100, Introduction to Engineering Design, provides students with a foundation for engineering design through hands-on team projects that address specified design opportunities.
Through this course, students will recognize the role that engineering and design have in improving the health, safety, and welfare of the global community, as well as identifying when a solution is technically feasible, economically viable, and desirable. Students will use a range of design tools and techniques to carry out and communicate their design processes as applied to their projects. Additionally, students will develop and practice professional skills, such as communication, teamwork, and ethical decision-making.

Course Learning Goals
discovered deep beneath the Pocono Mountains in Pennsylvania. The ore is located approximately 10,000 meters beneath the Earth's surface. However, due to the depth of the ore, mine infrastructure development proved to be extremely challenging, as air quality and temperature were difficult to control. Despite a large amount of ventilation infrastructure (including heating, ventilation, and air conditioning systems to cool the air), several mine workers died during development due to carbon monoxide poisoning or heat-related ailments.
Because of this, the Pennsylvania government has introduced new regulations regarding air quality for workers, which will make it impossible to use emissions-based equipment (e.g., diesel, natural gas) during the production phase of the mine. The mine was set up as a block cave operation and was intended to utilize load haul dump vehicles and haul trucks to extract and transport the ore to the surface for processing. Deviation or pursuit of a change from this mining method would be extremely costly and delay production for years. The Pennsylvania government has put out an open bid for an engineering company to develop a strategy to extract the ore cost-effectively and in the most environmentally-friendly manner. This could include vehicles with alternative power sources or new methods that could exploit the current mine setup. One government official stated special consideration would be given to bidders that could create jobs while ensuring worker safety with regard to environment (e.g., air quality, temperature) and general work hazards (e.g., crashes, cave-ins).
Each design team should research and develop a strategy to meet the bid objectives of the PA government. For your concept, consider alternatives to traditional mining methods and extraction equipment and provide recommendations with an emphasis on impact to: • Emissions/regulatory/environmental requirements, • Safety, • Costs (e.g., fuel, infrastructure), • Public opinion, and • Productivity.

Known parameters and assumptions
• The rare-earth element concentration is uniform throughout the ore deposit.
• The haul truck ramp from the extraction level to the surface processing plant is 8 km long at a constant 10% grade.
• The block cave draw points can provide 10 metric tons of fragmented ore per 10 metric tons extracted.
• There are 100 draw points in the mine.
• The average distance between a draw point and haul truck pickup locations is 300 meters.

Project deliverables
• A technical report containing the following elements: rationale for the recommendation, description of alternative concepts and their evaluation, systems diagram(s), concept of operations, environmental analysis, assessment of important aspects of your system for feasibility and adoption, including public opinion, and economic viability of the system.
• computer-aided drafting drawings, and • model or prototype of a component of the overall system."

Appendix C
Developed instrument to assess design project appropriateness for first year students in terms of STA, SCE, PDR, PGO and PAL. -.011 .815 -.051 -.128 .002 3. The skills I learned through this industrial sponsored project has helped me easily decide/choose/or stay with my major. 1. The project my group worked on was associated with a masculine or feminine experience (e.g., cooking makes me think of females).
.125 -.013 .810 .084 .013 2. The project my group worked on was associated with a masculine or feminine institution (e.g., the military makes me think of males).
-.088 -.076 .777 .053 .050 3. The project my group worked on was associated with a masculine or feminine action related (e.g., teaching makes me think of females).
.017 -.045 .841 -.020 .082 4. The project my group worked on was associated with a masculine or feminine interest related (e.g., war affects everyone, but men tend to be more interested).
.137 .017 .785 .133 .051 5. The project my group worked on was associated with a masculine or feminine idea generation (e.g., the ideas were mostly contributed by the males).
.064 .006 .734 .095 .016 6. The project my group worked on was associated with a masculine or feminine background knowledge (e.g., females know how to socially work in a group).