How to Use This App

  1. Log in to the MASDER Website

  2. Go to student surveys within the Administer (Surveys) tab

  3. Click on survey results

  4. Click on Download data to get a copy of your students responses

  5. Select the institutions, semesters, courses, and sections you would like to view from the checkboxes on the left.

Instructor Report

Welcome to your S-SOMAS Instructor Report! This report provides a summary of your students’ attitudes toward statistics, as measured by the S-SOMAS Survey. If your students completed the survey twice (such as in a pre/post format), you will see summaries of the attitudes and student characteristics at each time point separately, as well as a summary of the differences in attitudes across the two time points. If your students only completed the survey once, you will only see summaries of student characteristics and attitudes.

You will also see a comparison of your own course to our national sample of courses. This is to provide a reference point for you as you consider your results. The table of contents to the left can be used to quickly navigate this document.

Please note that you will not receive any identifiable data for your course. This report will include aggregated summaries of your course. You as an instructor should not be able to, but should also not attempt to, identify individual students based on individual responses in the report. To protect student identities, graphs are suppressed if they would show less than 5 responses. Deidentified data from all courses will be publicly released on the MASDER website in the future.

Overview of the MASDER Project

The S-SOMAS Survey is part of the MASDER (Motivational Attitudes in Statistics and Data Science Education) family of instruments. MASDER is a project focused on developing validated instruments to assess undergraduate student and instructor attitudes toward statistics and data science, and to measure characteristics of the learning environment. You can find more information at our website, https://sdsattitudes.com.

This material is based upon work supported by the National Science Foundation under Grant No. DUE-2013392. The opinions, findings, and conclusions or recommendations expressed are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Background of the S-SOMAS Survey

The Student Survey of Motivational Attitudes toward Statistics (S-SOMAS) is a tool intended to be used by educators and researchers for quantifying students’ attitudes about statistics containing 55 psychometric items. The items measure 8 constructs, and each item uses a 7-point Likert-type response scale (Strongly Disagree to Strongly Agree). Click here to see the survey. Note that the version that your students took may be slightly different than this one, depending on when your students took it.

In the statistics education literature, the term attitudes has historically been used to refer to affective constructs other than statistics anxiety. In development of the S-SOMAS, the research team adopted a widely-used psychological theory of motivation – Eccles’s Expectancy Value Theory (e.g., Eccles & Wigfield, 2020) – and coined the term motivational attitudes to bridge the existing statistics education literature with the EVT literature.

Theoretical Framework

Attitudes affect behaviours. Situated Expectancy-Value Theory (SEVT) is a popular way to conceptualize attitudes for research purposes. We chose SEVT as our theoretical framework because the attitude components it includes have been shown to predict student achievement choices and outcomes with similar populations. SEVT has also been used with college/university students in many disciplines (Wigfield & Eccles, 2020), including in the mathematics and statistics education literature (e.g., Meece et al., 1990; Ramirez et al., 2012; Simpkins et al., 2006; Unfried et al., 2015).

In SEVT, one’s achievement-related choices and performance are affected by what one values (Subjective Task Values) and what one expects to happen (Expectancies). All other constructs are mediated by the Expectancies and/or Subjective Task Values constructs, and a highly simplified representation of the SEVT model is shown in Figure 1.

With the S-SOMAS instrument, we measure 8 constructs which are described in the table below. The first four are Subjective Task Values. In this report, construct results from the S-SOMAS instrument are ordered according to this table.

Construct   # Items   Brief Definition
Academic Self-Concept   5   Student perceptions about the academic achievement (in general)
Attainment Value   5   How important success in learning statistics is to the student
Expectancy   4   How the student thinks they will perform in the field of statistics
Goal Orientation   4   What drives the students to learn statistics (might be internal or external reasons)
Interest/Enjoyment Value   4   The interest a student has in statistics, or their enjoyment from it
Utility Value   4   How useful the student perceives statistics to be for their personal life or career
Worth   4   How much time and effort the student perceives learning statistics to be worth
Cost Value   5   Factors that deter the student from learning statistics
Perception of Difficulty   4   How difficult the student perceives learning statistics to be

Data Collection

Your course was a part of our S-SOMAS Pilot 4 data collection cycle from Fall 2022 to Spring 2023. Instructors were recruited via listserv such as IsoStat, ASA SSDSE, and AMATYC. Only instructors teaching undergraduate introductory statistics courses in the United States were able to participate. Instructors who agreed to participate were given a Qualtrics link to the S-SOMAS for their course, and were responsible for administering the survey to their course at the beginning and/or the end of the course term. Instructors were encouraged to give course credit or extra credit to students who participated. This resulted in a convenience sample of 4,056 students in 77 sections with 27 instructors at 21 schools.

Data Cleaning

The S-SOMAS data were imported to R and cleaned before this report was created. The first step was to drop all respondents who did not consent to participate from the dataset; no data from individuals who do not consent is analyzed or included in the dataset. Additionally, all responses to Likert-type items were recoded as numeric, where Strongly Disagree is a 1 and Strongly Agree is a 7.

Matching Pre and Post Data

If your class completed both the S-SOMAS both pre-term and post-term surveys, the two datasets needed to be matched. We matched the pre/post data based on student email address. If an email address did not have a match, we attempted to rectify issues and matched responses manually based on First and Last name. We generated a variable that records for each student if they have Pre only, Post only, or Pre and Post data. In the report, a student’s data is included in any time point for which they had data.

Construct Scores

The psychometric items on the S-SOMAS are written so that they align to the 8 constructs shown in the table above. To create a score for each construct, the average of an individual’s responses to the items for that construct is computed. For the specific items that need to be reverse-coded prior to calculating the construct score, we recode so that Strongly Disagree is a 7 and Strongly Agree is a 1. We only calculate a construct score for a student if they answered at least 70 percent of the items on that construct.

How You Might Use These Results

These results are meant to help you to understand your students better and/or to understand yourself as an instructor better For example, a better understanding of your students’ attitudes may inform choices you make about your course pedagogy. We encourage you to consider both the magnitude of the attitude construct scores for your class as well as the score relative to the national sample. Here are some questions to guide your reading of the report:

  • In what areas do my students have high/positive attitudes?
  • In what areas do my students have low/negative attitudes?
  • In what areas are my students like the national sample? In what areas are they different?
  • How do the score results match what I expected to see?

Also consider if students show a change in attitudes over the course term. Note that large changes in attitudes within a single semester are generally not seen in statistics education research. Do not anticipate extreme changes in attitudes, but do pay attention to what changes you do see, and how those changes might relate to your course materials and/or teaching.

This report might also guide plans for professional development as you consider ways you might improve student attitudes. You might also choose to include results from this report in promotion or retention materials.

Note: This report provides information on a snapshot of students’ attitudes and possibly information about changes in students’ attitudes if appropriate data were collected. There are many ways that this information could be productively used, such as developing a better understanding of one’s students or evaluating a change in teaching approach such as a new curriculum or other intervention. These results may also be appropriate to consider as one part of the larger narrative of an instructor’s promotion or retention materials. However, we do NOT recommend these results be used directly in decisions regarding retention or promotion. (For example, we do NOT recommend that an instructor be terminated or not renewed because of the results shown in this report. There are many factors that contribute to students’ attitudes, and this report is NOT a direct evaluation of the instructor or their teaching.)

How to Interpret These Results

A construct score is provided for each construct; this is a value between 1 and 7 where lower values indicate that the respondent exhibits lower levels of the construct, and higher values indicate that the respondent exhibits higher values of the construct. For example, a student with a construct score of 6.2 for Utility Value would be said to have a high Utility Value for Statistics; that is, this student has positive attitudes about the usefulness of statistics. Note that the terms lower and higher are not normative and do not imply that a high score is necessarily better than a lower score. For example, a student with a scale score of 1.7 for Perception of Difficulty would have a low perception of the difficulty of learning statistics; that is, this student would think that learning statistics is easy. However, perceiving learning statistics as “easy” is not better than perceiving learning statistics as “difficult” (or vice versa).

In addition to construct scores, you will see results reported for individual survey items as well. Seeing these individual items can be very interesting, but for making more general conclusions about your students’ attitudes, we recommend focusing on construct scores as summaries of the 8 constructs.

How to Cite These Results

If you want to cite the S-SOMAS survey, please use the following:

MASDER Research Team. (2024). Student Survey of Motivational Attitudes toward Statistics (S-SOMAS). https:/masder.net

If you are interested in sharing these results publicly, please contact us before doing so.

Sharing your results internally is okay. If you do share your results internally, we recommend the following citation for your instructor report:

MASDER Research Team. (2024) S-SOMAS Instructor Report. https:/masder.net

For more papers and presentations about the development of these materials, please see our website https:/masder.net

Contact Us

Questions? Please contact us at masder.team@gmail.com, or visit our website at https:/masder.net

Construct Summary

This graph shows, for each construct, a comparison of the distribution of construct scores in your class compared to the national sample on the Pre survey. Dashed lines indicate mean values for each group. For each construct except for Perception of Difficulty, a higher score is considered a more positive attitude.

Loading...

Academic Self-Concept

Definition: Student perceptions about the academic achievement (in general).

Loading...

This graph shows the distribution of individual responses on each item within the Academic Self-Concept construct, comparing your class to the national sample on the Pre survey.

Attainment Value

Definition: How important success in learning statistics is to the student.

Loading...

This graph shows the distribution of individual responses on each item within the Attainment Value construct, comparing your class to the national sample on the Pre survey.

Expectancy

Definition: How the student thinks they will perform in the field of statistics.

Loading...

This graph shows the distribution of individual responses on each item within the Expectancy construct, comparing your class to the national sample on the Pre survey.

Goal Orientation

Definition: What drives the students to learn statistics (might be internal or external reasons).

Loading...

This graph shows the distribution of individual responses on each item within the Goal Orientation construct, comparing your class to the national sample on the Pre survey.

Interest/Enjoyment Value

Definition: The interest a student has in statistics, or their enjoyment from it.

Loading...

This graph shows the distribution of individual responses on each item within the Interest/Enjoyment Value construct, comparing your class to the national sample on the Pre survey.

Utility Value

Definition: How useful the student perceives statistics to be for their personal life or career.

Loading...

This graph shows the distribution of individual responses on each item within the Utility Value construct, comparing your class to the national sample on the Pre survey.

Worth

Definition: How much time and effort the student perceives learning statistics to be worth.

Loading...

This graph shows the distribution of individual responses on each item within the Worth construct, comparing your class to the national sample on the Pre survey.

Cost

Definition: Factors that deter the student from learning statistics.

Loading...

This graph shows the distribution of individual responses on each item within the Cost construct, comparing your class to the national sample on the Pre survey.

Perception of Difficulty

Definition: How difficult the student perceives learning statistics to be.

Loading...

This graph shows the distribution of individual responses on each item within the Perception of Difficulty construct, comparing your class to the national sample on the Pre survey.

Construct Summary

This graph shows, for each construct, a comparison of the distribution of construct scores in your class compared to the national sample on the Post survey. Dashed lines indicate mean values for each group. For each construct except for Perception of Difficulty, a higher score is considered a more positive attitude.

Loading...

Academic Self-Concept

Definition: Student perceptions about the academic achievement (in general).

Loading...

This graph shows the distribution of individual responses on each item within the Academic Self-Concept construct, comparing your class to the national sample on the Post survey.

Attainment Value

Definition: How important success in learning statistics is to the student.

Loading...

This graph shows the distribution of individual responses on each item within the Attainment Value construct, comparing your class to the national sample on the Post survey.

Expectancy

Definition: How the student thinks they will perform in the field of statistics.

Loading...

This graph shows the distribution of individual responses on each item within the Expectancy construct, comparing your class to the national sample on the Post survey.

Goal Orientation

Definition: What drives the students to learn statistics (might be internal or external reasons).

Loading...

This graph shows the distribution of individual responses on each item within the Goal Orientation construct, comparing your class to the national sample on the Post survey.

Interest/Enjoyment Value

Definition: The interest a student has in statistics, or their enjoyment from it.

Loading...

This graph shows the distribution of individual responses on each item within the Interest/Enjoyment Value construct, comparing your class to the national sample on the Post survey.

Utility Value

Definition: How useful the student perceives statistics to be for their personal life or career.

Loading...

This graph shows the distribution of individual responses on each item within the Utility Value construct, comparing your class to the national sample on the Post survey.

Worth

Definition: How much time and effort the student perceives learning statistics to be worth.

Loading...

This graph shows the distribution of individual responses on each item within the Worth construct, comparing your class to the national sample on the Post survey.

Cost

Definition: Factors that deter the student from learning statistics.

Loading...

This graph shows the distribution of individual responses on each item within the Cost construct, comparing your class to the national sample on the Post survey.

Perception of Difficulty

Definition: How difficult the student perceives learning statistics to be.

Loading...

This graph shows the distribution of individual responses on each item within the Perception of Difficulty construct, comparing your class to the national sample on the Post survey.

No post data available.

Construct Summary

This graph shows, for each construct, a comparison of the distribution of the change in construct scores over time in your class compared to the national sample. Dashed lines indicate mean difference (Post-Pre) values for each group. For each construct except for Perception of Difficulty, a score above 0 indicates a positive change in attitudes over the course of your term.

Loading...

Academic Self-Concept

Definition: Student perceptions about the academic achievement (in general).

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Attainment Value

Definition: How important success in learning statistics is to the student.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Expectancy

Definition: How the student thinks they will perform in the field of statistics.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Goal Orientation

Definition: What drives the students to learn statistics (might be internal or external reasons).

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Interest/Enjoyment Value

Definition: The interest a student has in statistics, or their enjoyment from it.


The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Utility Value

Definition: How useful the student perceives statistics to be for their personal life or career.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Worth

Definition: How much time and effort the student perceives learning statistics to be worth.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Cost

Definition: Factors that deter the student from learning statistics.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

Perception of Difficulty

Definition: How difficult the student perceives learning statistics to be.

Loading...

The "y=x" line represents students (or classes) where the construct pre-score is the same as the construct post-score.

The regression lines are generally flatter, having slopes less than 1. This indicates that students who get high scores on the pre-survey tend to decrease their scores on the post-survey. Conversely, students who get low scores on the pre-survey tend to increase their scores on the post-survey. (This is called regression to the mean).

When comparing the regression line for your class (in blue) to the national average (in red), we encourage you to focus on the difference in y-values. If the line for your students is above the national average, that indicates that your students show a higher increase in construct scores from pre-survey to post-survey. The difference between the slopes for class and national indicates that an increase is not the same for students with different pre-scores, which is generally less important.

No post data available.

Gender

Question wording: What is your gender?
Loading...

Race/Ethnicity/Origin

Question wording: What is your race, ethnicity, or origin? Select all that apply.
Loading...

Age

Question wording: What is your age? Please enter an integer.
Loading...

Class Level

Question wording: Which of the following best classifies you as a student?
Loading...

Previous Statistics Class

Question wording: Have you successfully completed a statistics class PRIOR to this semester / quarter?
Loading...

Enrollment Reasons

Question wording: Which of the following best describes why you are taking this class? *Select all that apply*.
Loading...

College GPA

Question wording: Please give your cumulative college GPA. For example, "My cumulative college GPA is 2.87 out of 4.0."
Loading...

Notes on College GPA data cleaning: Many students provided answers on GPA or GPA Scale which were not consistent with common GPA scales. As a result, we took the following steps in the data cleaning process:

  • If students reported that their College GPA was between 4 and 6 (including 6), we truncated it to 4.0
  • If students reported that their College GPA was 0, we set it to missing

Majors

Question wording: Pick the field that best describes your major or intended major.
Loading...

Minors

Question wording: If you have a minor, pick the field that best describes your minor.
Loading...