Measuring Leadership

MEASURING LEADERSHIP CA 670

In this course, we seek to study EXEMPLARY LEADERSHIP as defined by Kouzes and Posner: (a) Modeling the way, (b) inspiring a shared vision, (c) challenging the process, (d) enabling others to act, and (e) encouraging the heart. Being a supervisor, administrator, or manager does not mean the person is a leader the way we use the term. In fact, few people are exemplary leaders. Yet each person can be an exemplary leader because the role has nothing to do with position, and everything to do with behavior.

The goal of the course is for you to measure yourself against truly exemplary leaders and research-based information. Then create a plan and implement the plan to make yourself an exemplary leader in the five areas so you model, inspire, challenge, enable, and encourage.

Contact information: Dr. Joan E. Aitken, joan.aitken@park.edu Professor, Communication Arts, Park University, 229 Copley, 8700 NW River Park Drive, Parkville, MO 64152; 816-584-6785 (office).

COURSE MATERIALS

WELCOME TO MEASURING LEADERSHIP!

I hope you enjoy this course as much as I do. Don't hesitate to contact me. If you don't hear from me promptly, I may have missed your message. I receive so much nasty junk mail, be sure to put the course number CA670 in your subject line, so I will notice your email.

With warm regards,

Joan Aitken, Professor, Park University <Joan.aitken@park.edu 816-584-6785 (w) and 816-569-3566 (h)>

COURSE ACCESS Through eCollege

http://parkonline.org/

COURSE SYLLABUS

http://www.park.edu/syllabus/List.aspx

COURSE PLANNING WEBPAGE

Sometimes students have difficulty finding information in all the different eCollege categories. Thus, I provide all the course planning information located in ecollege is on one page for easy searching: http://onlineacademics.org/CA670/ To search, use �Control F� and enter the keyword. If you want to work ahead, you will be able to find course content here.

Required Textbook for Dr. Aitken's Course

TEXTBOOKS

Please obtain your required textbooks, skim the book, and read chapters one and two to begin. Used copies and student sharing should work fine. If you are having difficulty finding course materials, you can see what is needed at the Online Academics Bookstore, select "Measuring Leadership" located on the left of the page at http://onlineacademics.org/Store/

1. Kouzes LEADERSHIP PRACTICES INVENTORY. Required Leadership Assessment Inventory you will complete. No substitutions in Dr. Aitken's course. Get the Facilitator's Package that contains all the elements needed. You will need:

The Leadership Challenge, 3rd or 4th (hardback or softback) Edition by James M. Kouzes and Barry Z. Posner. Jossey-Bass
� ISBN-10: 0787984914
� ISBN-13: 978-0787984915

The Leadership Practices Inventory (LPI): Self Participant's Workbook with Self Insert (Package), One 120 page Participant's Workbook plus a 4 page Self Insert (The Leadership Practices Inventory) (Paperback)
by James M. Kouzes
� ISBN-10: 0787956562
� ISBN-13: 978-0787956561

Leadership Development Planner , 3rd Edition by James M. Kouzes and Barry Z. Posner. Jossey-Bass
� ISBN-10: 0787945684
� ISBN-13: 978-0787945688

The Leadership Practices Inventory - Scoring Software Rev 2e CD (CD-ROM)
by JM Kouzes (Author) � Publisher: Jossey Bass Wiley; 2nd ed edition (November 21, 2000)
� ISBN-10: 0787954128
� ISBN-13: 978-0787954123

The least expensive way to purchase may be through The Leadership Practices Inventory (LPI)-Deluxe Facilitator's Guide Package (Loose-leaf, with CD-ROM Scoring Software, Self/Observer, Workbook, Planner & copy of The Leadership Challenge book ) (3rd ed.) by Kouzes and Posner. Jossey-Bass, 2003. Apparently, this is not in print anymore, so you may want to find a used copy or buy the pieces individually.

2. Hackman, M. Z., & Johnson, C. E. (2004). Leadership: A Communication Perspective (4th or latest ed.). Long Grove, IL: Waveland Press. I have provided notes from the book on the course webpage. You probably used this textbook in another course. We will use this book as background and concentrate on the measures you can complete for this course.

--------------------------------------------------------------------

Recommended for Dr. Aitken's Course:

3. ENRON ETHICS CASE STUDY. Please view a documentary or read about Enron. Suggested sources are as follows:

ENRON: Smartest Guys in the Room Video or DVD.

Bryce, R. (2004). Pipe dreams: Greed, ego, and the death of Enron. New York: PublicAffairs.

--------------------------------------------------------------------

PREPARATION ASSIGNMENT

Begin the core assessment project week one by printing off and distributing the inventories to people you know (e.g., friends, coworkers, supervisors). Follow the LPI instructions, obtain 20 measures from co-workers and managers, and use the LPI software to generate results. Include the software report in your project. You may want to provide the entire project in a txt or rft document instead of a Word document (doc.).

The main thing is to not make people feel coerced into completing the inventories. "I'm doing this for a leadership class, and I need honest feedback so I can improve my communication skills. Please don't put your name on the form, and you can simply put it in this envelope with others. . . " Something along that line.

The measures are in the Facilitator's Package as leaflets (8 x 10). You can download the measure from here: Other Measure for the one you pass out and Self Measure for the one you complete on yourself. These materials are in the LPI facilitators package you purchased for the course. The scale is 1 (never) to 10 (always). Blanks and N/A should be recorded as a 1. The instructions are in your LPI package.

FURTHER INVESTIGATION

You may want to read online information about statistics, tests, and measurement. A useful source is Heffner, (2003). Research methods. All Psych Online. Contents - http://allpsych.com/researchmethods/researchcontents.html

RESEARCH FOR FURTHER INVESTIGATION

Peer-Reviewed Journal Article Research
You will want to conduct your own research to find scholarly journal articles to supplement your readings in this course. You may use this research in your weekly discussion and for your course project. Please use peer-reviewed journals in the field of communication studies.

Conduct your research in EBSCO's Communication and Mass Media Complete http://www.park.edu/library/ Your professor may suggest readings you also can access through EBSCO. You need to have the PDF reader software on your computer, so install Adobe PDF reader. http://www.adobe.com/products/acrobat/readstep2.html This is a free and safe download.

Dr. Aitken's Library Database Tutorial, click here.

Example Peer Reviewed Research Articles, which may be of interest for your LPI reports (Core Assessment).

Aldoory, L., & Toth, E. (2004). Leadership and gender in public relations: Perceived effectiveness of transformational and transactional leadership styles. Journal of Public Relations Research, 16(2), 157-183.

Becker, J., Halbesleben, J., & O'Hair, H. (2005). Defensive communication and burnout in the workplace: The mediating role of leader�member exchange. Communication Research Reports, 22(2), 143-150.

Boies, K., & Howell, J. (2006). Leader�member exchange in teams: An examination of the interaction between relationship differentiation and mean LMX in explaining team-level outcomes. Leadership Quarterly, 17(3), 246-257.

Campbell, C., & Swift, C. (2006). Attributional comparisons across biases and leader-member exchange status. Journal of Managerial Issues, 18(3), 393-408.

Campbell, K., White, C., & Johnson, D. (2003). Leader-member relations as a function of rapport management. Journal of Business Communication, 40(3), 170-194.

Douglas, C. (2006). Communication in the transition to self-directed work teams. Journal of Business Communication, 43(4), 295-321.

Fix, B., & Sias, P. (2006). Person-centered communication, leader-member exchange, and employee job satisfaction. Communication Research Reports, 23(1), 35-44.

Greguras, G., & Ford, J. (2006). An examination of the multidimensionality of supervisor and subordinate perceptions of leader-member exchange. Journal of Occupational & Organizational Psychology, 79(3), 433-465.

Kinnick, K., & Parton, S. (2005). Workplace communication. Business Communication Quarterly, 68(4), 429-456.

Lapierre, L., Hackett, R., & Taggar, S. (2006). A test of the links between family interference with work, job enrichment and leader�member exchange. Applied Psychology: An International Review, 55(4), 489-511.

Lewis, L. (2006). Employee perspectives on implementation communication as predictors of perceptions of success and resistance. Western Journal of Communication, 70(1), 23-46.

Limon, M., & France, B. (2005). Communication traits and leadership emergence: Examining the impact of argumentativeness, communication apprehension, and verbal aggressiveness in work groups. Southern Communication Journal, 70(2), 123-133.

McCroskey, J. (1992). Reliability and validity of the Willingness to Communicate Scale. Communication Quarterly, 40(1), 16-25.

Sharbrough, W. (2006). Motivating language in industry. Journal of Business Communication, 43(4), 322-343.

Teven, J., McCroskey, J., & Richmond, V. (2006). Communication correlates of perceived Machiavellianism of supervisors: Communication orientations and outcomes. Communication Quarterly, 54(2), 127-142.

Wolvin, A. (2005). Listening leadership: Hillary Clinton's Listening tour. International Journal of Listening, 19, 29-38.

WEBLINKS FOR FURTHER INVESTIGATION

Information on the latest scholarly developments in the field of leadership studies can be found in The Leadership Quarterly. This international journal publishes research and theoretical articles from many different disciplines, including anthropology, management, sociology, and political science. You can view a sample issue online at http://www.elsevier.com or request a subscription from The Leadership Quarterly, c/o Elsevier Science, Inc., 655 Avenue of the Americas, New York, NY 10010-5107.

To get in touch with other leadership educators, contact the International Leadership Association (ILA). An affiliate of the James MacGregor Burns Academy of Leadership at the University of Maryland, the ILA sponsors an annual conference, ongoing interest groups, leadership list serves, and a variety of leadership publications. A complete description of the association�s activities can be found here.

The Association of Leadership Educators (ALE) is another excellent resource. This group also hosts an annual convention and publishes The Journal of Leadership Education. For more information visit the ALE�s Web site here.

LECTURES BY WEEK OR UNIT

Week 1

MEASUREMENT

Week's Learning Outcomes:

a. Define exemplary leadership and culture. Identify basic communication and leadership research principles through the Hackman and Johnson textbook and other relevant materials (e.g., books, journal articles).

b. Define basic concepts of statistical analysis and measurement.

c. Measure and analyze your communication and leadership behaviors by completing measures found in the Hackman and Johnson textbook and the LPI.

Begin measuring your personal leadership.
Complete self-assessment measures.
Identify your leadership style.
Identify a successful leader you can interview and observe in action for the assignment due week 3.

d. Implement more effective exemplary leadership behaviors based on the Kouzes and Posner Five Principles and other course materials.

Identify what exemplary leaders do and what constituents expect.

Leadership Matters

Leadership is not a fad. It's a fact. It's not here today, gone tomorrow. It's here today, and here forever.

Leadership makes a difference. Try naming one significant movement that wandered leaderless into the history books. And leadership matters most in times of uncertainty. The study of leadership is the study of how men and women guide us through adversity, uncertainty, hardship, disruption, transformation, transition, recovery, and new beginnings. Challenge is the opportunity for greatness. Given the daunting challenges we face today, the potential for greatness is phenomenal.

People matter. Even in today's wired world, it's not the web of technology but the web of people that matters most. Leaders can't do it alone. Success in any project, organization, enterprise and in life has been, is now, and will continue to be a function of how well people work and engage with each other. Success in leadership depends on your capacity to build and sustain collaborative human relationships.

You matter. People who become leaders don't always seek the challenges they face. Challenges also seek leaders. It's not so important whether you find the challenges or they find you. What is important are the choices you make when stuff happens. The next time you say to yourself, "Why don't they do something about this?" look in the mirror. Ask the person you see, "Why don't I do something about this?" The legacy that you leave will be the life that you lead.

Descriptive Statistics

Directly quoted or adapted from Overton, T. (2006). Assessing learners with special needs: An applied approach. 5th ed. Upper Saddle River, NJ: Pearson. Chapter 3. (For more information, click here.)

Why is measurement important?
In educational contexts, the educational future of so many individuals is at risk. In business, decisions need to be based on the best possible information. Of concern are studies indicating typical types of mistakes made by professionals in the field regarding their misinterpretation of tests and data. Data may play little if any part in decision making. Professionals continue to support poor quality tests, when better tests are available. Common errors include:
1. Using instruments in the a assessment process solely because those instruments are stipulated by the organization or management.
2. Regularly using instruments for purposes other than those for which test have been validated.
3. Taking the recommended use at face value.
4. Using the quickest instruments available even though those instruments may not assess the areas of concern.
5. Using currently popular instruments for assessment.
6. Failing to establish effective rapport with the examinee.
7. Failing to document behaviors of the examinee during assessment.
8. Failing to adhere to standardized administration rules.
9. Making various scoring errors, such as
a. Making simple counting errors.
b. Making simple subtraction errors.
c. Entering data incorrectly, such as on the wrong line.

What are numerical scales?
Nominal scale consists of numbers used only for identification purposes (numbers that cannot be used in mathematical operations).

An example of a nominal scale is to name or classify "male" or "female."
Ordinal scale uses rank ( does not have the quality of equidistant units).
Interval scale are equidistant units (numbers that cannot be used in mathematical operations).
Ratio scale does have an absolute zero and can be used in mathematical operations.

What are descriptive statistics?
Descriptive statistics are used to organize and describe data. Derived scores are scores obtained by using a raw score and expectancy tables. Below is an expectancy table (source).

The image �http://www.internetraining.com/Statkit/SK30.gif� cannot be displayed, because it contains errors.

Standard scores are derived scores that represent equal units; also known as linear scores. Here's a great page with a calculator that will convert information to standard scores, click here.

The image �http://tusdstats.tusd.k12.az.us/planning/resources/dr_gregg/standardyl.jpg� cannot be displayed, because it contains errors.

Measures of central tendency are statistical methods for observing how data cluster around the mean. The mean is the average, which can be affected by an extreme score, especially if the group is composed of only a few people.

Protect yourself in communication research by understanding basics of statistics and measurement.

Normal distribution or the bell curve (below) is a symmetrical distribution with a single numerical representation for the mean, median, and mode.

Frequency distribution is a method of determining how many times each score occurs in a set of data.
Mode is the most frequently occurring score in a set of scores.

Just as measure of central tendency are used to see how sets of data cluster together around an average score, measures of dispersion are used to calculate how scores are spread from the mean. The way that scores in a set of data are spread apart is known as the variability of the scores, or how much the scores vary from each other. Variance describes the total amount that a group of scores varies in a set of data: The degree or amount of variability or dispersion in a set of scores.

Across our population, for example, we expect Intelligence Quotient (IQ) to have a normal distribution like this picture. On tests measuring intelligence, the mean IQ is 100. One hundred is also the middlemost score (median) and the most frequently occurring score (mode). In fact, more than 68% of all of the IQ scores will cluster within 1 standard deviation, or one determined typical unit, above and below the score of 100.

Notice the total percentages in each category (68% in the two middle categories and 96% in the four middle categories).

No need to feel humiliated if you don't already know this stuff. You'll get the hang of it

We also indicate the same information by standard deviations, as indicated below. The standard deviation helps the teacher determine how much distance from the mean is typical and how much is considered significant. The standard deviation of a set of data is the square root of the variance. More than 68% of the scores fall within 1 standard deviation above or below the mean. A normal distribution is symmetrical and has the same number representing the mean, median, and mode. Notice that approximately 95% of the scores are found within 2 standard deviations above and below the mean. A student within the 70-75 IQ range may be found eligible for services under the category of mental retardation if there are additional supporting data. I person with an IQ around 130 may be called "gifted."

When selecting norm-referenced tests, read the test manual and determine the size of the sample used in the norming process. Tests developed using larger samples are thought to result in scores that are more representative of the majority population.

Percentile ranks and z scores provide additional ways of looking at data. Using percentile ranks is a method of ranking each score on the continuum of the normal distribution. Percentile ranks are scores that express the percentage of people who scored as well as or lower than a given person's score. Percentiles range from the 99.9th percentile to less than the 1st percentile.

Some tests us T scores to interpret test performance. T score have an average or mean of 50 and standard deviation of 10.

No, a T score has nothing to do with being a tough-guy.

A z score indicates where a score is located in terms of standard deviation units. The mean is expressed as 0, 1 standard deviation above the mean is expressed as +1, 2 standard deviations above as +2, and so on. z scores are derived scores that are expressed in standard deviation units.

There is nothing wrong with getting prayer help for understanding z scores.

Stanines divide the scores into 9 groups of scores and are reported as 1 through 9 with a mean of 5. The standard deviation unit of stanines is 2. Deciles are scores that are reported in 10 groups ranging from a score of 10 for the lowest grouping to 100 for the highest group of scores. Each grouping represents 10% of the obtained scores.

You will see the same basic information also described through percentages, percentiles, Z scores, T scores, and Standard Nine (Stanines), as represented below.

Some sets of data have to modes or two most frequently occurring scores, which is known as a bimodal distribution. A distribution with three or more modes s called a multimodal distribution. A frequency polygon is a graphic representation of how often each score occurs in a set of data.


Frequency Polygon (Source)	Clearly, this guy thinks he's a polygon.

When small samples of populations are tested or when a fairly restricted population is tested, the results may not be distributed in a normal curve. Distributions can be skewed in a positive or negative direction. When many of the scores are below the mean, the distribution is said to be positively skewed.

Reliability and Validity

Directly quoted or adapted from Overton, T. (2006). Assessing learners with special needs: An applied approach. 5th ed. Upper Saddle River, NJ: Pearson. Chapter 4. (For more information on the topic, click here.)

Reliability in assessment refers to the confidence that can be placed in an instrument to yield the same score for the same person if the test were administered more than once and to the degree with which a skill or trait is measured consistently across items of a test.

Example of low reliability regarding watch dog protection.

Correlation is a statistical method of observing or calculating the degree of relationship between two sets of data on two variables. There are three types of correlations between variables: positive, negative, and no relationship. The degree of relationship between two variables is expressed by a correlation coefficient (r). The correlation coefficient will be a number between +1.00 and -1.00. r values of -.78 and +.78 are of equal strength.

There is usually a positive correlation for being a ferocious dog and eating baby bunnies.

There is usually an inverse or negative correlation between being a ferocious dog and being a baby bunny.

A scattergram is a graphic representation of a correlation. The closer the dots approximate a straight line the nearer to perfect the correlation.

What are methods of measuring reliability?
We must be able to have confidence that test instruments used will yield similar results when administered at different times. One correlational method used is the Pearson's Product Moment correlation, known as Pearson's r. Pearson's r is a commonly used formula for data on an interval or a ratio scale. Internal consistency is the consistency of the items on an instrument to measure a skill, trait, or domain.

Internal consistency is the consistency of the items on an instrument to measure a skill, trait, or domain.

When a respondent says he is a hot dog on one question and a giraffe on another question, the test has low internal consistency.

Test-retest reliability is a study that employs the administration of a single instrument to check for consistency across time.

When retaking a test, and individual may remember certain items. This practice effect most likely would cause the scores to be higher on the second administration.

If the dog says she is a blonde the first time she takes the test and a red-head the second time she takes the test, we suspect poor test-retest reliability.

Equivalent forms reliability AKA alternate forms reliability is consistency of a test to measure some domain, traits, or skill using like forms of the same instrument. The items are matched for difficulty on each test. During the reliability study, each person is administered both forms, and the scores obtained on one form of the test are then paired with the scores obtained on the equivalent form.

Split-half reliability is a method of checking the consistency across items by halving a test and administering two half-forms of the same test. Because most tests have the items arranged sequentially, from the easiest items at the beginning of the test to the most difficult items at the end, the tests are typically split by pulling every other item, which then allows for a single administration of the instrument.

These are not an example of split half reliability because these are actually two different dogs.

Kuder-Richardson (K-R) 20 is a formula used to check consistency across items of an instrument with right/wrong responses. An advantage is a single test administration.

When the subject's self concept suggests a princess fantasy, we can say the test has external validity in this case. The test reported what it purports to measure, and thus has external validity.

Coefficient alpha is a formula used to check consistency across terms of an instrument with responses with varying credit.

Coefficient alpha is not to be confused with alpha dog. Or maybe it is.

Interrater reliability is the consistency of a test to measure a skill, trait, or domain across examiners. This can be accomplished by administering the test and then having an objective scorer also score the test results. This information is especially important when tests with a great deal of subjectivity are used in making educational decisions.

An adequate reliability coefficient would be .60 or greater, and a high degree of reliability would be above .80. For example, if the examiner is interested in measuring a trait over time, the examiner should select an instrument in which the reliability or consistency over time had been studied.

The calculation of the reliability coefficient is a group statistic and can be influenced by the makeup of the group.

These two dogs have high inter-rater reliability because they both came up with the same answer.

What is standard error of measurement?
Standard error of measurement is the amount of error determined to exist using a specific instrument, calculated using the instrument's standard deviation and reliability. In all psychoeducational assessment, there is a basic underlying assumption: Error exists! Errors in testing may result from situational factors such as a poor testing environment or the health or emotions of the person being tested, or errors may occur due to inaccuracies within the test instrument. Error should be considered when tests are administered, scored, and interpreted. A single test score may not accurately reflect the student's true score. True score is the person's actual score. Obtained score is the observed score of a person on a particular test on a given day.

Obtained score = True score + Error

Obtained score - True score = Error

To estimate the amount of error present in an individual obtained score, the standard error of measurement must be obtained and applied to each score. The standard deviation and the reliability coefficient of the instrument are used to calculate the standard error of measurement.

Oh yes, Error exists!

Confidence interval is the range of scores for an obtained score determined by adding and subtracting standard error of measurement unit. Using the standard error of measurement to judge the test's quality is more important than reliability. Some testing practitioners fail to consider possible test error when interpreting test results of a person being evaluated.

Estimated true score is a method of calculating the amount of error correlated with the distance of the score from the mean of the group. The father from a test mean a particular person's score is, the greater the change for error within the obtained score. The estimated true score can be used to establish a range of scores by using the standard error of measurement for the estimate true score. All test scores contain error.

Definitely, there is test error in this case.

What is test validity?
Validity is the quality of a test; the degree to which an instrument measures what it was designed to measure. Validity is concerned with good results for the purpose of the test.

Criterion-related validity is the statistical method of comparing an instrument's ability to measure a skill, trait, or domain with an existing instrument or other criterion. This method compares its scores with other criteria known to be indicators of the same trait or skill that the test developer wishes to measure.

Concurrent validity is a comparison of one instrument with another within a short period of time, typically the same day.

Predictive validity is a measure of how well an instrument can predict performance at a later date on some other variable.

Content validity occurs when the items contained within the test are representative of the content purposed to be measured. For a test to have good content validity, it must contain the content in a representative fashion. Presentation format is the method by which items of an instrument are presented to an individual. Response mode is the method required for the examinee to answer items of an instrument.<