Self-efficacy is defined as a person's belief about his or her capability to perform at a certain level for a specific task,1 which affects how individuals approach tasks, how they respond to failure, and how they choose to use their knowledge and skills.2 Self-efficacy differs from general self-confidence because self-confidence is a more global concept indicating the strength of the individual's belief about his or her self-worth but is not related to a specific task as self-efficacy is. Self-efficacy also differs from self-assessment because self-efficacy uses an internal reference, whereas self-assessment uses an external criterion.3
“Individuals'” beliefs in their capabilities can affect the type and level of difficulty of goals they set and the strength of their commitment to them.4 The strength of perceived self-efficacy is not linearly related to the choice to attempt a specific behavior, and a minimal threshold of self-assurance is needed even to attempt any task or activity. Experiencing higher levels of self-efficacy results in the same willingness to attempt an activity, but the stronger the sense of self-efficacy, the greater the perseverance to complete the activity and the higher the likelihood that the chosen activity will be performed successfully.5 Therefore, student physical therapists with higher levels of self-efficacy in professional behaviors and in patient management skills should be more persistent and more successful in achieving the expected outcomes of graduate physical therapists.
In his seminal 1997 work, Self-Efficacy: The Exercise of Control,6 Bandura4 described a self-efficacy theory which has since been widely applied to fields including education, health, athletics, the corporate world, and social and political change. Self-efficacy is believed to provide a link between knowledge and the application of knowledge and skills to specific situations and is critical to the development of competence and expertise in health professions students.7
Tools That Measure Self-efficacy
Self-efficacy is not a global trait but is a set of self-beliefs linked to distinct areas of functioning; therefore, self-efficacy measures should be tailored to a selected domain.4 Task demands differ in difficulty and in a developmental hierarchy, that is, some tasks must be mastered before other tasks can be attempted. Self-efficacy assessments should measure behaviors over which the participants have some control.5 Because self-efficacy is context specific, there is no single valid standardized measure of perceived self-efficacy. Construct validation can be assessed by correlation with other valid measures so that individuals who score high on self-efficacy assessments should score high on a valid measure of ability in the same construct. Individuals who score high on measures of self-efficacy should differ from those who score low in comparison with another measure, for example, grades, performance assessments, or licensure outcomes.5
To ensure valid measures based on accurate conceptions of self-efficacy theory, Bandura5 published a guide for constructing self-efficacy measures. His suggestions included that individuals should be presented with items reflecting varying levels of task demands. The items should be reflective of current capabilities, not expectations for the future. For each item, participants rate their confidence for what they think they “can” do, not what they think they “will” do. Using the verb “can” reflects a measure of capability, whereas “will” indicates intention. Participants rate each item based on the strength of their belief in their ability to complete each activity described in the individual items.5 Rating scales range from 0, cannot do the activity at all, to 100, very certain they can do the activity. As in all other testing, to minimize response bias, the individuals should be assured that their responses are completed privately and confidentially.5 Self-efficacy measures in the health professions follow Bandura's guidance for context-specific criteria or statements, including a way to assess the individual's level of confidence, and the use of descriptors or anchors in the rating scale.7
Several self-efficacy scales for physical therapist education or practice have been documented in the literature.7–9 Venskus and Craig8 developed and validated the Physical Therapist Self-Efficacy (PTSE) scale, which consists of 5 statements focusing on a physical therapists' self-efficacy in clinical reasoning and 8 statements about general self-confidence, which is inconsistent with Bandura's definition of self-efficacy. The PTSE scale was based on current literature about the components of clinical reasoning in the management of a patient or client by a physical therapist (PT). The items measuring self-efficacy in clinical reasoning are general and not directed toward a specific context. For example, 1 item is “I am confident that I know when to perform specific tests for PT practice.”8 The authors validated the PTSE scale by analyzing scores across years of study in an academic curriculum, reasoning that self-efficacy would develop as students gained knowledge, skills, and experience.8
Jones and Sheppard7 developed a 13-item self-efficacy scale for physical therapist students based on criteria on which students would be assessed during an acute care clinical placement. The Physiotherapy Self-Efficacy (PSE) questionnaire includes 13 items, in which there were 2 items dealing with communication, 2 on completing an assessment, 2 items referring to interpreting the assessment, and 7 focusing on interventions, all related to an acute care setting. They used student feedback to establish face validity and found a moderate positive correlation between self-efficacy and preclinical grades (Spearman rho = .55).7 The PSE questionnaire has been translated into Dutch, and the Dutch version's 3 components of clinical management had excellent internal consistency (Cronbach's alpha > .90; Intraclass Correlation Coefficient [ICC] = .80).10 van Lankveld et al10 concluded that self-efficacy in student physical therapists is not generic but is specific to the clinical area of practice, for example, in an acute care setting.
Salbach et al9 developed and validated an 11-item self-efficacy scale regarding evidence-based physical therapist practice, the evidence-based practice confidence (EPIC) scale. The items in the scale represent steps relevant to evidence-based practice, an important construct thought to reduce unwarranted variation in practice. The EPIC's reliability and validity were studied using practicing physical therapists, not students.9 The researchers for the present study conducted an extensive review of the literature, and when it was determined that no existing survey satisfied a research question about Doctor of Physical Therapy (DPT) student self-efficacy and educational outcomes, the researchers developed the Self-Efficacy of Student Physical Therapist Outcomes survey for the purpose of filling that void.
Expected Outcomes of Entry-Level Physical Therapists
The Commission on Accreditation in Physical Therapy Education (CAPTE) defines expected outcomes as the “competencies that the program expects students to have achieved at completion of the program.”11 Grignonet al12 conducted a qualitative study identifying expected outcomes for physical therapist graduates that were grouped into 3 categories: communication skills, competence in patient care, and professional obligations, which were linked to CAPTE criteria.11 These expected outcomes were used as the theoretical foundation for the current study and formed the basis for the development of a self-efficacy scale for physical therapist practice. The purpose of this study was to establish the validity of the Self-Efficacy of Physical Therapist Student Outcomes (SEPTSO), the researcher-developed tool used in the study. The second purpose was to assess self-efficacy using the SEPTSO in students from a DPT program.
One hundred twenty-two graduate students enrolled in the Doctor of Physical Therapy program at Texas State University were invited to participate in the study. Students in 3 cohorts (DPT 1, DPT 2, and DPT 3) of the 9-semester (3-year) graduate program were included in the study. There were no exclusion criteria. The DPT curriculum consists of 6 and one-half semesters of academic coursework. During either the fifth or sixth semester, students participate in an integrated clinical education experience in which they are responsible for patient management in an on-site clinic. This part-time clinical experience consists of 75 hours of practice. Academic coursework is followed by 2 and one-half semesters (32 weeks) of full-time clinical experiences.
This study was a methodological research using a longitudinal, observational design in which participants completed a survey 3 times within 1 academic year. The study was approved as exempt from oversight by the Texas State University's Institutional Review Board (2016H9820).
Outcome Tool Design and Development
The SEPTSO survey was designed by 2 of the researchers (L.A.S., J.R.B.) according to the guidelines established by Bandura.5 One researcher (J.R.B.) has extensive experience in assessment of and interventions to enhance self-efficacy and served as a content expert in developing the survey. The SEPTSO was developed based on the outcomes expected of physical therapist graduates identified by Grignon et al,12 which were service and social responsibility, professionalism, professional role, professional commitment, practice management, communication, professional growth and development, evidence-based practice, clinical reasoning, and patient management. The study survey also included 7 demographic questions about the participants' sex, age, personality style, educational background, and previous career experience.
Each of the 25 items was reviewed by 2 of the researchers (L.A.S., S.S.) for face validity using Grignon's 10 expected outcomes, definitions, and example behaviors. Each researcher independently identified the expected outcome that was most closely reflected in each of the 25 items. For example, “Professionalism” was defined by Grignon et al12 as adhering to conduct that is aligned with defined ethical and moral obligations and responsibilities and example behaviors included demonstrating the core values of the profession, practicing patient-centered care, exhibiting cultural competence, and practicing legally. The researchers independently linked SEPTSO items 2, 3, and 4 to Professionalism. When disagreements about the items occurred, they were resolved by the third researcher (J.R.B.).
Jette and Portney13 validated a model of 7 professional behaviors for student physical therapists. While interpreting the findings for the present study, the researchers realized that many of the items in the SEPTSO items were representative of Jette and Portney's13 7 professional behaviors. One of the researchers (L.A.S.) reviewed the behavioral descriptions of the professional behaviors and, using definitions and example behaviors by Grignon et al,12 identified the SEPTSO items that were representative of Jette and Portney's13 professional behaviors. A second researcher independently identified SEPTSO items that could be aligned with Jette and Portney's13 professional behaviors using the same methodology. Any disagreements were resolved by the third researcher (J.R.B.). For example, the behaviors described in “Professionalism” include self-presentation, accountability, integrity, and professional values.13 Grignon et al12 defined professionalism as “adhering to conduct that is aligned with defined ethical and moral obligations and responsibilities” and gave example behaviors of demonstrating the core values of the profession, practicing patient-centered care, exhibiting cultural competence, and practicing legally. The SEPTSO items that were based on Grignon et al's12 professionalism definition included the following items: 1) Practice ethically in all situations, 2) Practice legally in all situations, and 3) Practice in a culturally sensitive manner using concepts of patient-centered care. These 3 items also met the behaviors described by Jette and Portney13 for “Professionalism.”
Content validity is concerned with demonstrating that an instrument adequately represents all content defining the variable being measured.14 SEPTSO items were based on the theoretical construct of the expected outcomes of physical therapist graduates.12 Some outcomes were represented by more than 1 item to better define them in behavioral terms. For example, patient management was separated into 5 areas of practice defined by the Guide to Physical Therapist Practice: musculoskeletal, neurological, cardiopulmonary, integumentary, and pediatrics.15 The stem question for each item was “How confident are you that you can ____ right now?” Response choices began at 0% and increased in 10% intervals to the maximum of 100% (Appendix A, Supplemental Digital Content 1, http://links.lww.com/JOPTE/A65).
Construct validity provides evidence to support the theoretical framework and the degree that an instrument reflects all theoretical components of the construct.14 Part of construct validity is based on content validity; that is, the ability to define clearly the theoretical context of an instrument. Construct validity is difficult to demonstrate because it reflects the ability of an instrument to adequately measure an abstract concept, such as self-efficacy. One method of addressing construct validity is with a factor analysis.14
The SEPTSO was composed of 25 items that together should reflect the characteristics of self-efficacy. A measure with good internal consistency should reflect homogeneity of each item in the scale, which is typically measured using Cronbach's alpha.14
Participants were invited to complete a 32-item survey, 7 demographic questions plus the SEPTSO, at the beginning of the Summer semester, which corresponds to the beginning of the program year. The online survey was delivered to each student by email through a link delivered by SurveyMonkey, and students participated by completing the survey anonymously. A reminder email was sent to students 1 week after the initial invitation to participate. All members of the 3 cohorts were invited to participate at the beginning of the next 2 semesters, Fall and Spring, whether they had submitted the survey in the first administration or not. Therefore, all students in each of the 3 cohorts were invited to fill out the SEPTSO 3 times.
Descriptive analysis was used to identify frequencies of the participants' demographic data, which were gathered in the first 7 items of the survey. Answers to each self-efficacy item in the SEPTSO reflect the percent of confidence identified by the participants for each specific activity. For data analysis, an average total score for each participant was developed by weighing the percent scores from 1 to 11 with 0% confidence weighted as 1, 10% confidence weighted as 2, and through 100% confidence weighted as 11. The weighted frequencies were averaged for each cohort including a grand total frequency for all participants.
Several analyses were used to assess the content validity of the SEPTSO to represent the theoretical components of self-efficacy in the expected graduate outcomes construct. A 1-factor Analysis of Variance (ANOVA) was used to determine whether there were differences among the 3 cohorts of student physical therapists' self-efficacy for expected behaviors of graduates at the first administration of the SEPTSO. A second 1-factor ANOVA was used to identify differences among all participants' averaged scores over the 3 administration times. In this case, a 1-factor ANOVA was used rather than a repeated measures ANOVA because the participants were not matched across all 3 survey administrations and different students participated in different administrations of the survey. Ages of participants were grouped into 4 categories: 21–24, 25–29, 30–34, and more than 35 years. To determine whether there were differences among sex (woman, man) and age groups' self-efficacy scores during the Summer administration of the SEPTSO, a 2 × 4 ANOVA was used. To determine the content validity of each item, a 3 × 25 ANOVA was used to determine differences among the 3 cohorts for each SEPTSO item.
Construct validity was determined using an exploratory factor analysis with a principal component analysis with varimax rotation to identify any meaningful components and to explain variance within the items. The threshold cutoff factor for including an item in a factor was .4, as recommended by Portney and Watkins.14 Finally, a Cronbach's alpha was used to determine internal consistency using 198 data points gathered from all individual responses in the 3 cohorts in each of the 3 testing periods.
Of the 3 cohorts enrolled in the Doctor of Physical Therapy program invited to participate in the study, 72 students in the Summer semester participated with a return rate of 59%. Of these 72 respondents, 19 were DPT 3 students, 26 were DPT 2 students, and 27 were DPT 1 students. The respondents included 24 men and 48 women, and the mean age was 26.33 years (range 21–40 years). Of the 72 students who responded, all had earned a bachelor's degree and 8 held an earned master's degree. Fifty-seven percent of previous degrees were in exercise and sports science or kinesiology. Physical therapy was a second or third career for 24 students, and several of these students reported more than 1 career, resulting in 38 previous careers. In the Fall and Spring administration, all 122 students from the target sample size were invited to participate, whether they had completed the first round or not. In the Fall, 62 students participated, with a 51% return rate, and in the Spring administration, 65 students participated, with a 53% return rate. In the second and third survey administrations, a technical error allowed the participants to complete the survey without identifying their class, resulting in the inability to determine the number of students in each class. Because different students participated in the 3 different administration times and because responses were anonymous, individuals were not followed up across time.
Twenty-four of the 25 items in the SEPTSO corresponded to at least one of the expected outcomes identified by Grignon et al12 No SEPTSO item was linked to the outcome “Professional Growth and Development.” Two outcomes corresponded to only 1 SEPTSO item: “Service and Responsibility” corresponded to SEPTSO #1, and “Professional Commitment” corresponded to SEPTSO #9. The other 7 outcomes were linked to 3 or more SEPTSO items (Table 1). Although Jette and Portney's13 professional behavior model was not used to develop the SEPTSO, the close connections between their model and the items in SEPTSO add to the strength of the face validity of the SEPTSO. One professional behavior, Personal Balance, was linked to only 1 SEPTSO item: 16) Receive feedback in an open manner to enhance performance. All other professional behaviors were represented by multiple SEPTSO items (Table 2).
The total mean weighted self-efficacy scores (range 0–100) for each class at the Summer administration were DPT 1 = 32.30 (SD = 14.59), DPT 2 = 67.42 (SD = 6.30), and DPT 3 = 73.88 (SD = 8.67). A 1-factor ANOVA identified a statistically significant difference among the 3 cohorts in the first administration of total average scores .05F74 = 51.93, ρ < .0001. A Tukey's HSD revealed significant differences at α = .05 between self-efficacy in the classes of DPT 1 and DPT 3 (ρ < .001) and DPT 2 and DPT 3 (ρ < .001).
The total mean self-efficacy scores for each testing period including all 3 student classes were Summer = 55.23 (SD = 22.53), Fall = 72.22 (SD = 15.65), and Spring = 80.49 (SD = 21.42). A 1-factor ANOVA identified a statistically significant difference among the 3 testing periods in self-efficacy for expected behaviors of graduates .05F2 = 32.21, ρ < .001. A Tukey's HSD found significant differences between self-efficacy in Summer and Fall (ρ < .001), Summer and Spring (ρ < .001), and Fall and Spring (ρ = .035). A 2 × 4 ANOVA resulted in no significant differences for the main effects of age (ρ = .059) or sex (ρ = .113) and no significant difference in their interaction (ρ = .843).
Item Construct Validity
A 3 × 25 ANOVA identified a significance difference (ρ < .05) for the individual item scores among each cohort. A Tukey's HSD found significant differences between DPT 1 and 2 cohorts and between DPT 1 and 3 cohorts (ρ < .05) for each item in the SEPTSO with the following exceptions. There were significant differences among all 3 cohorts (ρ < .05) in SEPTSO: #11 Delegate patient care appropriately, #12 Supervise PTA and Aides, #21 Provide safe, effective, and efficient patient care for a patient with a musculoskeletal disorder, #23 Provide safe, effective, and efficient patient care for a patient with a cardiopulmonary disorder, and #24 Provide safe, effective, and efficient patient care for a patient with an integumentary disorder. Two items were significantly different only between the DPT 1 and 3 cohorts: #16 Receive feedback in an open manner to enhance performance (ρ = .004) and #2 Practice ethically in all situations (ρ = .001). Finally, there were significant differences between the DPT 1 and 3 cohorts (ρ = .000) and the DPT 2 and 3 cohorts (ρ = .000) for item #25) Provide safe, effective, and efficient care for a pediatric patient (Table 3).
Internal consistency was excellent with Cronbach's alpha = .983 for the 25 items. Finally, the inter-item correlation was moderate to good (.51–.75) and good to excellent (>.75) for most correlations. All correlations were significant at ρ < .05. See Supplemental Digital Content 2 (Appendix B, http://links.lww.com/JOPTE/A66).
An exploratory factor analysis with a principal component analysis and varimax rotation resulted in a model in which 2 factors explained 71% of the variance within the items. A threshold cutoff of less than or equal to .04 was used to identify loading factors. Component 1 Patient Management (eigenvalue 16.19) accounted for 65% of the variance and included 14 of the 25 items in the SEPTSO. Component 2 Professionalism (eigenvalue 1.64) accounted for 7% of the variance and included the remaining 11 items (Table 4).
Based on the psychometric evaluation performed in this study, the SEPTSO demonstrates excellent content validity, responsiveness, and lack of bias for sex and age. Internal reliability (Cronbach's alpha = .983) was found to be excellent, indicating that all items represent the same construct of self-efficacy. The content validity of the SEPTSO was good because each item could be directly linked to the expected outcomes identified by Grignon et al12 and the practice patterns identified in the Guide.15 The SEPTSO survey was able to differentiate between the first, second, and third year DPT students and in scores across the 3 testing periods demonstrating the responsiveness of the survey. DPT 3 students' self-efficacy scores were higher than DPT 1 and DPT 2 students' scores, indicating that perceived self-efficacy likely increases as students get closer to graduation. This finding is not surprising because self-efficacy typically increases as behaviors become habitual and students accumulate more clinical practice as they get closer to graduation, essentially creating clinical habits. The longitudinal finding in the present study extends the findings of Venskus and Craig8 who identified that self-efficacy related to clinical decision-making is greater in third year DPT students compared with the second and first year DPT students and in the second year DPT students compared with the first year DPT students. This finding demonstrates that students' confidence in their capability to demonstrate knowledge and skill increases as they progress through the program, providing a strong foundation for developing into autonomous practitioners. The finding that DPT 1 and DPT 2 students' perceived self-efficacy scores were not different is postulated to be related to the amount of clinical practice experienced. Students in the program do not participate in full-time clinical education experiences until the third year of the curriculum, so both DPT 1 and DPT 2 students had not participated in full-time clinical education. The role of clinical education in enhancing perceived self-efficacy of DPT students should be explored in future studies.
An exploratory factor analysis with a principal component analysis and varimax rotation identified 2 factors, or subscales, in the survey. The patient management factor explained 71% of the variance, and the professionalism factor explained 7% of the variance. All survey items fit into 1 of the 2 factors. Patient management requires both clinical knowledge and skills to effectively treat a patient and an understanding of other issues that make up physical therapist practice.16 The patient management factor in this study asked the participants to rate their confidence to provide safe, effective, and efficient patient care for patients with a variety of disorders categorized by a practice pattern, such as musculoskeletal practice. The patient management factor also included components directly related to providing evidence-based patient management, providing patient education, and included administrative aspects of patient care such as supervision of physical therapist assistants and technicians, and documentation.16
Self-efficacy is an important construct to assess in student learners, especially learners such as physical therapists whose craft includes psychomotor skills and behaviors in which repetition and practice are required to achieve competence.17 In addition to influencing the performance of behaviors, self-efficacy affects other aspects of human functioning, such as goals and aspirations, outcome expectations, the perception of benefits and barriers in the social environment, optimistic versus pessimistic thinking, and the pathways individuals pursue. Given that a goal of DPT education is to produce mature, emotionally intelligent, high-functioning physical therapists who can effectively adapt in a complex environment, self-efficacy is a construct that educators should assess and monitor. The SEPTSO demonstrated strong psychometric properties in the measurement of the construct of perceived self-efficacy of DPT students and therefore can be used in physical therapist education programs to assess self-efficacy in student learners. Sequential measurement of self-efficacy across the curriculum (eg, each semester or every 6 months) could identify individual students who may need additional practice to enhance confidence and thus performance, an application of the SEPTSO that was not assessed in the current study and that should be tested in future investigations. The SEPTSO survey could be used to inform student self-reflection, because of the relationship between self-efficacy and aspects of human functioning, providing additional value in DPT education programs. Programs can use serial measures of self-efficacy as an outcome measure in the context of accreditation, and program and curriculum evaluation.
The SEPTSO survey has many strengths that may set it apart from the other tools designed to measure self-efficacy in DPT students. The SEPTSO survey was constructed in a manner consistent with Bandura's5 guidelines, resulting in items worded in the present tense (“How confident are you right now?”) and rated on a 100-point scale. The PSE questionnaire7 and the PTSE8 do not adhere to Bandura's guidelines. Both questionnaires use a 5-point response scale, which Bandura5 indicates is less sensitive and reliable because respondents typically avoid the 2 extreme positions (1 and 5), leaving only a few options (2, 3, and 4). The 100-point scale provides more differentiating information and is therefore a stronger predictor of performance. Furthermore, Bandura5 recommends that self-efficacy scale response items be worded such that the items refer to confidence “today” or in the present. The PSE7 and the PTSE8 contain several items that are future oriented rather than present oriented, which may be an assessment of a person's potential rather than their actual ability. The authors believe that the SEPTSO survey is a more accurate measure of self-efficacy because it adheres to Bandura's guidelines for self-efficacy scale construction.
Education can influence a person's self-efficacy, so specific learning activities should include components designed to enhance self-awareness and self-efficacy; however, the knowledge and skills learned should be equivalent to the amount of self-efficacy gained.7 For example, Jones and Sheppard3 found a strong, negative correlation (Spearman rho = −.72) between self-efficacy and clinical performance in student physical therapists who participated in 8 hours of simulation. These students overestimated their performance, that is, their self-efficacy was greater than their actual ability.3 Bandura6 has identified 4 types of intervention approaches that enhance self-efficacy, including vicarious or observational learning, social and verbal persuasion, drawing attention to physiological and affective states, and the facilitation of repetitive but achievable performance accomplishments or skills mastery. Some of these approaches are currently incorporated into DPT education programs, and they are familiar to physical therapists because they use them to develop skills mastery in patients. For example, students often begin their learning process of a new skill by watching the instructor demonstrate the skill (vicarious learning). Additional strategies to enhance self-efficacy may be novel to DPT educators. Debriefing with students after the practice or application of a new skill using questions such as “what was the best thing about this experience?” and “what did you learn from this experience?” are examples of social and verbal persuasion. Focusing on the positives in a novel experience has been found to enhance self-efficacy, even in situations in which the negatives outweigh the positives.18 Creating short-term learning goals that are both challenging and achievable will enhance self-efficacy, whereas setting goals that are too easy or too difficult and thus not achievable serve to decrease self-efficacy (an example of skills mastery). Incorporating opportunities in the classroom or during the DPT program to process, acknowledge, and affirm how students are feeling as they develop competency and build confidence in their knowledge and skills is an example of drawing attention to physiological and affective states. Based on the strong relationship between self-efficacy and performance that has been documented across multiple different behaviors, educators should consider intentionally adopting strategies to enhance student self-efficacy.
A suggestion for future research is to apply the SEPTSO survey to multiple DPT programs to determine its generalizability and to determine whether students' self-efficacy correlates with ratings on a clinical education assessment tool or with passing the licensure examination. The SEPTSO survey also could be used as a component of program evaluation to indirectly assess the program's curriculum. Another suggestion for future research is to use the SEPTSO survey to identify areas needing improvement for individual students to help them successfully complete their professional education. For example, students with lower self-efficacy scores related to patient management with an integumentary disorder would benefit from additional tutoring or practice in skills such as evaluation or debridement of a wound. Likewise, if a student is demonstrating difficulties in one of the professionalism behaviors, such as providing or receiving feedback, additional opportunities to practice these skills may make it more likely that the student will be successful in a clinical setting.
Individual students were not tracked across time, which would have provided the ability to correlate the SEPTSO survey with other measures of achievement, like GPA, course grades, and clinical performance, providing additional information about the usefulness and role of assessing self-efficacy in DPT education. Only students from 1 DPT program were included, so the results of this study may or may not be generalizable to students in other physical therapist education programs.
The SEPTSO survey demonstrated excellent content validity, responsiveness, and internal reliability without bias for age or sex. An exploratory factor analysis identified 2 factors, patient management, which explained 65% of the variance, and professionalism, which explained 7% of the variance. All 25 items fit into 1 of the 2 factors. The potential uses of the SEPTSO survey include the following: to identify program weaknesses in program evaluation, to identify targets for enhancement in curriculum evaluation, to serve as an outcome measure for accreditation, and to encourage self-reflection in students. At a minimum, incorporating approaches to enhance student self-efficacy appears to be a sound educational strategy.