May 4, 2009 Self-Identification of LGBT Employees P a g e | 1

A living report generated by the LGBT Self-ID Community of Practice http://lgbtselfid.collectivex.com/

May 6, 2009

Since the 1990s, thousands of employers both in the United States and abroad have improved their employment policies, benefits and other internal and external practices to ensure more equitable treatment of their lesbian, gay, bisexual, and transgender (LGBT) employees, although laws explicitly protecting workers based on gender identity or sexual orientation are inconsistent across the United States and internationally.

Anecdotal evidence supports these policies, benefits, and practices as recruitment, development, and retention tools for LGBT employees; however, little empirical data exists to support this. Since gender identity and sexual orientation are generally hidden, visibly indistinguishable characteristics in the workplace, evaluating the success of these practices is difficult because most employers do not have a sense of how many LGBT employees they have or where in their businesses LGBT employees actually work. Organizations that have committed to equal treatment of LGBT workers — through non-discrimination policies, employee benefits and training programs — should evaluate recruitment, retention and productivity of LGBT workers. Having business metrics to quantitatively evaluate these programs is critical to a viable, fully inclusive diversity program.

This report is the result of the collaborative effort of almost 80 participating professionals representing more than 35 U.S. and Canadian employers and nonprofit organizations. The group drew from its own expertise and ultimately interviewed eight major U.S. employers to prepare recommendations for businesses seeking to implement self-identification programs. These recommendations are presented in the full report, including specific examples from the employers interviewed.

There are two primary methods of capturing employees’ gender identity or sexual orientation: Anonymous Engagement Surveys and Confidential Employee Records. Employers may opt for one or both, depending on their current level of commitment to LGBT workplace equality. Both methods can help illuminate organizational deficiencies affecting employee recruitment and retention, but each offers particular advantages and limitations.

Existing engagement and climate surveys can include optional demographic questions on gender identity and sexual orientation. Participation will likely be greater when the collection method clearly assures the confidentiality of the responses and the anonymity of the LGBT employee.

By including questions about gender identity and sexual orientation in the same breath as questions about race, ethnicity, and gender, the survey approach can positively signal the importance of equity for all employees regardless of these non-performance-related characteristics.

However, when engagement surveys are conducted anonymously, the data collected has more limited use for talent and workforce management purposes than more direct Human Resources Information System (HRIS) approaches. Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 2

Employers that have consistently and comprehensively sent a strong organizational message of LGBT inclusion and demonstrated organizational competency with LGBT employees may opt to expand human resource information and data collection systems to include gender identity and sexual orientation identification data with employee records along with EEOC data on gender, race, and ethnicity. This approach requires the development of measurement criteria which allows monitoring on an on-going basis. Clearly communicating intended uses of the information is critical to LGBT participation – e.g., relate to promotions, leadership, and coaching opportunities.

Some employers may wish to implement a ―phased release,‖ beginning with at least two business units in the first year and expanding to other business units after the first year. Others may consider an opt-in approach, such as inviting employees to participate in a standalone LGBT employee engagement survey which may subsequently be recorded in the employee records.

To drive employee participation and success of these programs over time, employers must consider several specific aspects in collecting the required data:

U.S. employers with 100 or more employees and certain federal contractors are required to provide the Equal Employment Opportunity Commission an annual count of their employees by job category and then by ethnicity, race and gender.1 Military and disability status are also commonly reported. Employers typically use such reporting to self-assess progress towards internal hiring and other diversity benchmarks. Thousands of employers both in the United States and abroad have improved their employment policies, benefits and other internal and external practices to ensure more equitable treatment of their lesbian, gay, bisexual and transgender (LGBT) employees, although laws explicitly protecting workers based on gender identity or sexual orientation are inconsistent across the United States and internationally. Although anecdotal evidence supports these policies, benefits and practices as recruitment, retention and productivity tools for LGBT employees, little empirical data exists to support this. Employers have largely not tracked the gender identity and sexual orientation of their employees — characteristics vital to fully understanding the diversity of a particular workforce. In the context of the work environment, both gender identity and sexual orientation are generally hidden, visibly indistinguishable characteristics. As a result, evaluating the success of these practices is difficult because most employers do not have a sense of how many LGBT employees they have or where in their businesses LGBT employees actually work. Approximation methods such as membership in LGBT employee groups or enrollment in partner benefits programs are less than ideal in comparison to a parallel form of data collection as is used for other characteristics because LGBT employees may not participate in these programs.

1 http://www.eeoc.gov/eeo1survey/faq.html and http://www.eeoc.gov/eeo1/eeo1_2007_d.pdf

Efforts to identify and quantify the experiences of LGBT employees date at least as far back as 2001, when Bank of America first gave LGBT employees the option of self-identifying in anonymous employee engagement surveys.2 More recently, employers have begun confidentially collecting individual employees’ gender identity and sexual orientation data to support targeted recruitment, retention and development programs for LGBT employees. However, businesses gathering this data largely developed proprietary guidelines and controls for collecting LGBT identification data that are not transparent or easily reproducible for other businesses. In 2006, a group of more than 80 interested LGBT-identified professionals, including human resources, workforce management and employment law professionals and LGBT workplace advocates set out to identify best practices for employers to collect gender identity and sexual orientation data from employees in optional and anonymous or confidential mechanisms. The group drew from its own expertise and ultimately interviewed eight major U.S. employers to prepare recommendations for businesses seeking to implement self-identification programs, including guidance to address common privacy concerns and drive employee participation over time. These recommendations are presented in this report, including specific examples from the employers interviewed. Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 4

Organizations that have committed to equal treatment of LGBT workers — through non-discrimination policies, employee benefits and training programs — should evaluate recruitment, retention and productivity of LGBT workers. Having business metrics to quantitatively evaluate these programs is critical to growing and expanding a viable diversity program. Employers typically capture an employee’s race, ethnicity, gender and sometimes military and disability status to quantitatively evaluate recruitment and retention across the entire organization and within individual business units. It makes good business sense to leverage existing data collection methods to also track and gain a more complete picture of LGBT employees to significantly enhance efforts of diversity/inclusion employee resource groups, task forces and other initiatives.

―The makeup of today’s workforce is becoming more diverse as a result of demographic changes. Since this makeup is continuously being altered, it is necessary for the individuals responsible for implementing and leading diversity-related initiatives to constantly monitor trends and themes relating to diversity.‖ - SHRM 2006 Workplace Diversity and Changes to the EEO-1 Process Survey Report3 ―With a ‘non-visible’ minority such as [LGBT workers], it used to be difficult to give managers real numbers of employees impacted to help them make important business decisions. Now that we have actual numbers, those discussions are much easier. We can tell managers that X number of employees in their site will be affected by a decision. Additionally, and perhaps most importantly, senior leaders of the firm receive results of their business unit’s performance across all key measures. This has helped us to identify specific areas for improvement across many of the diversity criteria... Allowing LGBT employees to self-identify... enabled the firm to better understand and respond to the needs and experiences of LGBT employees relating to satisfaction with the organization, retention, growth opportunities [and] inclusion.‖ - J.P. Morgan Chase & Co. ―So far, we have collected only limited data. Even so, we have found value in the data itself, as well as in the process of collecting it. In terms of the numbers, we learned that we had more transgender employees than previously assumed (~0.5% of the responding populations). In terms of the results, we found that, overall, GLB and T employees were at least as satisfied as the average employee. In addition, one difference was identified that we plan to watch in future surveys.‖ - Corning Inc.

3http://www.shrm.org/hrresources/surveys_published/2006%20Workplace%20Diversity%20And%20Changes%20To%20The%20EEO-1%20Process%20Survey%20Report.pdf – SHRM.org paid membership required to access Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 5

There are two primary methods to capture employees’ gender identity or sexual orientation. Employers may opt for one or both, depending on their current level of LGBT competency:

Both methods can help illuminate organizational deficiencies affecting employee recruitment, retention and productivity, but each offers particular advantages and limitations. An organization’s overall cultural and LGBT-specific competency will help determine which method the organization is ready to introduce. Employers should consider the following before engaging in self-identification methods:

When were gender identity and sexual orientation added to the non-discrimination policy? Are employees aware of the policy?

Does the organization have a culture where senior executives have consistently or openly talked about gender identity or sexual orientation?

Has the employer ever asked any employees questions about their gender identity or sexual orientation?

To more fully assess LGBT competency, an organization may use tools such as ―The 15 Steps of an Out & Equal Workplace.‖4 Indications of moderate competency may suggest the use of engagement surveys as the first step, while indications of strong competency may suggest using engagement surveys and employee records. The initial program should be voluntary for LGBT employees until such issues as the climate for LGBT employees who self identify can be determined.

4 http://outandequal.org/steps-to-equal-workplace Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 6

Existing engagement and climate surveys can include optional demographic questions on gender identity and sexual orientation. Participation will likely be greater when the collection method clearly assures the confidentiality of the responses and the anonymity of the LGBT employee, including assurances that:

An employee will not be identifiable by drilling down the reported data – for example, Corning Inc. does not report results for a particular demographic within any business unit where fewer than five individuals identify as a member of that demographic; Bank of America Corp. assures its employees that ―only those statistical groupings with at least 20 responses will be reported to ensure your anonymity.‖

Results will be collected and examined by a third-party vendor;

Results will have limited authorized access by the employer; and

Specific verbatim responses that provide identifiable information will not be reported or will be redacted to protect the individual.

By including questions about gender identity and sexual orientation with questions about race, ethnicity and gender, the survey approach can positively signal the importance of equity for all employees regardless of these non-performance-related characteristics.

However, when engagement surveys are conducted anonymously, the data collected has more limited use for talent and workforce management purposes than more direct HRIS approaches.

If employees choose to self-identify, the data collected enables an employer to:

Analyze and compare demographics of information collected to establish an average and variances from the average;

Identify where greater investment or attention may be needed (e.g.: where input is below average);

Identify more effective activities where input is above average;

Leverage leadership on the perceptions and real input from their LGBT workforce for recruiting, development and retention purposes;

Identify where the business might be at risk for legal discrimination or hostile work environment concerns if input is overly negative;

Identify trends in comfort levels for openly identifying as LGBT in the workplace -- a lack of responses or uncharacteristically low presence in a business unit might indicate a perceived hostile work environment which can be reviewed by rank, geographic location or business unit;

Enable satisfaction assessment and issue recognition by location or department organization with the ability to ―cut‖ input by LGBT identity and/or ―out‖ status; and

Solicit input for improvements, effects of efforts on particular considerations, or overall efforts.

Employers that have consistently and comprehensively sent a strong organizational message of LGBT inclusion and demonstrated organizational competency with LGBT employees may opt to expand human resource information and data collection systems to include gender identity and sexual orientation identification data with employee records along with EEOC data on gender, race and ethnicity.

Existing workforce management and individually-based career development programs can be expanded to include employees based on gender identity and sexual orientation. Data for existing employees can be collected as part of routine updates of personal information, and data for new employees should be collected at the same time as race, ethnicity and gender data are collected.

Workforce management personnel should be familiar with LGBT-specific considerations when addressing productivity, career development and other opportunities with LGBT employees.

Ongoing analysis of collected LGBT employment data is essential to fully realizing the benefits of this approach. Furthermore, clearly communicating intended uses of the information — e.g.: leadership and coaching opportunities— is critical to employee participation in the long term.

If employees choose to self-identify, the data collected enables analysis and accountability of LGBT measures such as:

Reviewing hiring, retention and involuntary separations of individual LGBT employees;

Reviewing individual LGBT employees for pipeline and succession purposes;

Identifying role models to support mentoring and individual development of individual LGBT employees;

Analyzing performance rating and promotion differentials;

Providing developmental opportunities such as candidates for inclusiveness speaking engagements;

Enhancing representation of non-visible minorities in processes like recruiting.

Some employers may wish to implement a self-identification program in a phased release, beginning with at least two business units in the first year and expanding to other business units after the first year. Alternatively, this may be implemented by geography; many companies consider starting in the U.S. and then expanding to other countries.

Others may begin more simply by collecting and maintaining data through LGBT employee groups, but such data should not be considered representative of a business’ larger LGBT workforce because many LGBT employees won’t feel comfortable joining an employee group and the group may not be present throughout all areas and levels of the business’ operations.

Regardless, make sure that any phased release or additional survey includes questions that appear in other organizational surveys for comparison and benchmarking purposes.

Employers may consider an opt-in approach, such as inviting employees to participate in a standalone LGBT employee engagement survey which may subsequently be recorded in employee records. Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 8

Careful, deliberate question wording is critical as perceptions and understanding of the terminology and question construction could potentially discourage or encourage an LGBT person from disclosing, or have non-LGBT employees incorrectly identify as LGBT. Work environments where employee LGBT identity is not openly discussed or where the culture does not welcome open discussion, may signal a more reserved choice of wordings, as well as a more limited number of potential questions.

When drafting these questions, please consider the following:

Provide separate questions for gender identity and sexual orientation — everyone has a gender identity and a sexual orientation, but the two are distinct, unique characteristics even though the community is known collectively as LGBT;

Make the questions optional, allowing employees to bypass them without answering, if they choose;

Provide the employer’s diversity and inclusion statement prior to asking any demographic questions, including questions about ethnicity, race and gender, to reinforce messages about confidentiality and employee privacy;

Provide space for employees to express their identity in their own words. Look for ways to fully encompass all aspects of a category – e.g. heterosexual/straight, gay, bisexual or all genders including transgender options; and

Ask about an employee’s level of being open about their gender identity and sexual orientation (e.g. workplace vs. personal life; team vs. organization; management vs. co-workers) to identify trends and deficiencies.

Questions about gender identity and gender, sex or sexual orientation should be separated. Whereas the concept of gender relies on cultural and social norms, sex refers to biological characteristics. However, the EEOC does not clearly delineate between sex and gender, and laws governing an individual’s ability to change their legal sex marker on government-issued identification vary from jurisdiction to jurisdiction within the U.S. The following questions are recommended options to clearly identify transgender people:

Is your gender identity different from the sex you were assigned at birth? (Yes/No)

Are you / do you identify as transgender? (Yes/No)

Gender identity can conceptually be more nuanced and fluid than simply ―female‖ or ―male.‖ The above questions may be followed up with additional questions to determine how transgender employees prefer to be identified and whether or not they are open about their transgender identity:

What best describes your gender identity?

o Female

o Male

If you identify as transgender, are you open about your gender identity: Yes Partially No

a. At home

b. With colleagues

c. With your manager

d. At work generally

The following questions were derived from the Stonewall Monitoring Report. The first question can be used independently or in conjunction with the second question.

1. What is your sexual orientation?

a. Bisexual

b. Gay man

c. Gay woman/ lesbian

d. Heterosexual/ straight

e. Other __________________

f. Prefer not to say

2. If you identify as lesbian, gay, bisexual or other, are you open about your sexual orientation: Yes Partially No

e. At home

f. With colleagues

g. With your manager

h. At work generally

In engagement surveys, employers may want to limit the number of questions being asked of all employees, for example when a general survey has additional LGBT-specific questions that will only be asked of people who identify as LGBT such as satisfaction with benefits, level of being ―out‖ or ―open‖ at work, and so on.

The two employers interviewed for this report that include LGBT identity in confidential employee records, IBM and an anonymous employer (see methodology), both asked a single question relating to LGBT identity, similar to the following.

1. Do you identify as lesbian, gay, bisexual or transgender? (Yes/No)

Four of the eight employers interviewed for this report utilized a single, broader ―LGBT identity‖ question as opposed to separate questions specific to gender identity and sexual orientation. It is unclear whether using a single question is advantageous or disadvantageous compared with separate questions, but some LGBT people may not generally associate with the broader LGBT community and their information might be better captured with separate questions.

In engagement surveys, some employers take the ―select all that apply‖ approach for demographic questions, which ask the employee to choose among a wide array of demographic questions. Corning Inc. uses the following:

Lesbian, gay or bisexual

Transgender

At the request of its employees, Sun Microsystems now allows employees to self-identify and designate information to acknowledge their same-sex spouse in communications, although this information ―does not feed or inform any of [Sun’s] benefits systems.‖ Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 10

To drive employee participation and success of these programs over time, employers must acknowledge the limitations of LGBT identification. To overcome these limitations, employers must work to ensure privacy and confidentiality of LGBT employees and employ consistent communications throughout the program.

Regardless of which method of identification an employer chooses, it should acknowledge the limitations posed by LGBT employees that will not feel comfortable sharing their identity with their employer.

―The idea of self ID through the associate survey was vetted with our Pride Resource affinity group. The first year, some associates expressed concern[s] that the information may not be kept confidential. Concerns have significantly lessened now that associates have taken the survey for a few years and understand information will be used only in aggregate and kept confidential. The demographic items are [still] optional on the survey.‖ – Anonymous Employer (see methodology) Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 11

Because employers have typically not asked employees to identify based on gender identity or sexual orientation, and because protections for LGBT employees are not uniform between jurisdictions, many LGBT employees will be concerned about disclosing this information. In engagement surveys, aggregated reports specific to smaller business units or functions could potentially expose LGBT employees. On employee records, the individual LGBT employee’s perception of the business’ competency about and commitment to LGBT inclusion may raise concerns about the use and privacy of the data.

Methods for collecting information about the gender identity and sexual orientation of employees lack the relatively strong framework that generally exists for other characteristics. Whichever method is leveraged, the employer must be prepared to address privacy and legal issues. A self-identification program must make LGBT identification questions optional by providing the answer option ―I do not want to disclose‖ or clearly indicating how to skip the question – whether or not employees answer LGBT identification questions is useful data to evaluate, particularly over time.

Some employees may be uncomfortable sharing this information at all, or having it associated with their employee records, especially if they are not open to anyone at work about their gender identity or sexual orientation. While some employees may never share such information, participation will likely grow as employees gain trust in the process over time. Participation is expected to be greater when employers clearly communicate:

Why the data is being collected;

Where the data will be stored;

Who will have access to the data; and

How the information will be used and reported.

Respecting personal privacy is paramount — the employer is obliged to preserve the confidentiality of personal information. Employers should consider both local and global policies pertaining to personal data collection and dissemination.

Employees must clearly understand what data are anonymous and what will be associated with their names and employee records. Once collected, data that is stored needs to be identified to adhere to privacy and confidentiality purposes. For example, if data is stored in the employee’s records, and non-essential personnel have access to that employee’s records, such personnel should not be able to view the particular data. Alignment with current processes and policies is essential to help streamline and support the new option.

The consent of the employee must be obtained for the collection, use or disclosure of personal information, except where not required to do so by law. The purposes for which information is collected used or disclosed by any employer must be those that a reasonable person would consider appropriate in the circumstances, and should be disclosed before employees are given the option to self-identify. Before an organization uses personal information that has been collected for a purpose not previously identified, it should identify the new purpose for the employee to whom the personal information relates prior to using the information in that manner except as permitted or required by law. The form and manner of obtaining consent may vary from express written consent to implied consent, depending upon the circumstances and the type of information. In determining the form and manner of consent, the organization should take into account the sensitivity of the information and the reasonable expectations of the employee. Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 12

Successful implementation of either method requires a cohesive communications and implementation strategy. Communications should at a minimum explain the reasons for the new data collection and the intended purposes of the effort compared to other existing efforts targeting ethnic, racial, gender or other characteristics. Strong communications will help build trust necessary for employees to feel comfortable disclosing their gender identity and sexual orientation over time. Implementation planning should leverage existing processes for collecting this data, such as on-boarding of new hires or annual updating of personnel information.

These communications should not be limited to LGBT self-identification, and should be a part of a larger diversity and identification program that drives towards maximum participation and accurate data about all characteristics, including ethnic, racial, and gender status of the workforce, to demonstrate the way this data can work together to improve workplace equality.

Whatever approach an employer ultimately takes to gather data on workers’ gender identity and sexual orientation should be communicated to all workers in a thoughtful manner. Virtually all of the employers interviewed for this report engaged existing LGBT employee groups to build trust and develop a common understanding of the goals and intentions of the program, although this would not necessarily reach LGBT employees that have not joined the group. Communications about LGBT data collection should be addressed to the entire workforce, creating open dialogue throughout the business and emphasizing senior management commitment to the project.

As part of the data collection tool, the employer could include a brief note to explain why questions about these characteristics have been added and reaffirm the employer’s commitment to fair treatment of all workers.

Anonymous Engagement Survey Introduction Language for Demographic Data Questions:

Each of us is unique. With this in mind, please indicate below how you would describe your own uniqueness. Your answers are voluntary, confidential and will be used only to better understand patterns of diversity across the Technology Community. - Corning Inc.

Feedback from employees tells us that it is important to address diversity in the broadest sense, beyond race and gender. One way to achieve this is to better understand those aspects of diversity that are not immediately obvious. With this in mind, please check off any of the following categories that you want to include in your responses. Your answers are voluntary, confidential and will be used ONLY to understand and address issues that may exist in our business. - J.P. Morgan Chase & Co.

Confidential Employee Record Introduction Language for Demographic Data Questions:

EMPLOYER recognizes the sensitivity of this information and takes every precaution to protect its confidentiality. Thus, this information will only be used for governmental and internal diversity reporting and other legitimate business purposes. EMPLOYER is required by U.S. federal regulations to maintain, track and report information only on the race or ethnicity and gender of the U.S. workforce. The information that you furnish will not be used as the basis for any employment action, and it will not subject you to any adverse treatment. - Anonymous Employer (see methodology)

An employer must be prepared to respond to data collected through engagement surveys or employee records, including data indicating who skipped or did not answer LGBT identification questions.

Analyze data for disparity in hiring, promotion and satisfaction levels across business units.

Be prepared to engage specific units in LGBT competency training.

―We used the [first year’s] survey data to get ideas on what we should be focusing on nationally... We used [the second year’s survey data] to compare against some of our benchmarks and see where we improved... We were able to see where we had our out gay leadership more clearly and where we needed to build more awareness. There was clear data on who needed improvements in comparison to other geographies and practices.‖ - Ernst & Young LLP

Although there are privacy and data usage concerns, none of the employers surveyed in this effort encountered legal concerns that would prohibit self-identification through either method, as long as it is voluntary and data is properly safeguarded.

In the event that survey responses reveal potential discrimination or harassment concerns such as derogatory comments, some employers provide these responses to their in-house counsel.

As with any type of engagement survey that includes demographic data (e.g.: race, ethnicity, gender), organizational leaders need to be prepared to take corrective action should the data reveal that discrimination is occurring.

Legislation to add gender identity and sexual orientation as protected categories to federal employment non-discrimination law in the United States was introduced in 2007 and will be reintroduced in 2009. The proposed legislation explicitly precludes the Equal Employment Opportunity Commission from collecting data based on gender identity or sexual orientation. No employers surveyed in this effort expected any changes to their self-identification method should such legislation be enacted as law.

The global nature of modern businesses presents some additional concerns for LGBT self-identification efforts, particularly because certain countries cannot provide a safe harbor for personnel who do choose to self-identify.

The best practices outlined in this report are based on U.S. guidelines and current status. Both Bank of America Corp. and J.P. Morgan Chase & Co. indicate that their anonymous surveys include this question in worldwide operations.

Global considerations such as Canadian or European data collection polices may need to be considered if changes in either the nature of data collected via surveys or any additions to HR systems are introduced. IBM has encountered strong resistance to tracking LGBT identity through confidential employee records in certain countries. Businesses must consider how to handle individual employee records for employees who are transferred to countries without protections for LGBT employees and their families.

For the above reasons, some companies have implemented data collection in the United States first, expanding to other countries over time. Where are our LGBT employees? May 4, 2009 Self-Identification of LGBT Employees P a g e | 14

General trends: For example, if the data reveals that LGBT employees are being laid off at a higher rate than other groups.

If an employer is committed to equal opportunity, having these data will enable them to further this goal. If the data reveal inequalities, management must be prepared to take action. Taking action to correct inequities can ultimately prevent lawsuits in the future. The data alone do not prove discrimination; the data may be coincidental.

If you are going to do tracking, management should be committed to doing adverse impact analyses for any layoffs. Once an employer has data they need to be committed to using the data to prevent inequality and discrimination.

How have you convinced management to take this risk?

What other risks did you uncover?

How have you seen this or used this as a useful tool in your company?

What other benefits have you discovered?

How did you decide that this was a good thing to do?

Having the data could also be useful for protections in the case to mitigate risk ahead of time.

Use a third party to collect the data for surveys to maintain anonymity.

Use a dedicated server and databases with strong access controls to limit who can view employee identified answers.

Develop contingency plans to address data leaks according to local laws and corporate policy. Provide notification of employee should data leaks occur.

One possible recommendation would be to store this information separately from the employee’s personnel file. Perhaps treat this as confidential medical records are treated today.

Transgender employees may be covered under HIPPA; therefore, these data need to be treated with the same privacy policies as other medical information.

Country and state laws will vary. Employers will need to conduct a legal review to ensure data privacy procedures are in compliance. Employers should be able to leverage procedures and policies for other types of personal privacy data.

How does your organization ensure confidentiality?

How have you addressed management’s concern of increased exposure due to perceptions of potential retaliation, discrimination, and harassment?

How have you addressed legal fears of having data that could be used in class action lawsuits?

What if someone answers honestly, and the self-ID information is accidently leaked or data security is breached (for example, laptop is stolen with confidential information). What risk mitigation plans have you put in place?

Can you explain how you have implemented the sexual orientation and gender identity program?

Use standard data access policies for private data. For example: use firewalled or dedicated servers and databases.

One strategy is to send a general climate survey to all employees and then ask demographic questions that apply to all employees. For example, gender, age, race, sexual orientation, gender identity, etc. Additional questions can be asked based on responses to these questions. Ultimately, the data can be cross-tabulated to report differences under various demographics.

Specific questions could be asked around sexual orientation and gender identity as well.

Use your LGBT employee group to gauge their opinions about LGBT specific issues.

Leverage the LGBT employee group members to communicate with the company of issues of self-identification and workplace climate.

What techniques have you used to get the climate survey out to LGBT employees?

Do you include a demographic question on the survey that allows people to Self-ID as LGBT or reveal their sexual orientation and gender identity?

Have you limited your climate survey to LGBT employees only?

Have you done similar climate surveys for other minorities?

In 2005, Howard Solomon of Sun Microsystems attended a session at the Out and Equal Conference that covered the need of viable tracking of LGBT employees for recruitment and retention purposes. With approval from Sun, Howard reached out to several LGBT employee groups at other businesses across a variety of industries.

From there, a group of interested LGBT-identified professionals — including human resources, workforce management and employment law professionals and LGBT workplace advocates — joined together to identify best practices for employers to collect gender identity and sexual orientation data from employees in optional and anonymous or confidential mechanisms.

To prepare recommendations for businesses seeking to implement self-identification programs, including guidance to address common privacy concerns and drive employee participation over time, the group drew from its own expertise and ultimately interviewed eight major U.S. employers:

Bank of America Corp.,

Corning Inc.,

Ernst & Young LLP,

Hewitt Associates Inc.,

IBM Corp.,

J.P. Morgan Chase & Co.,

Sun Microsystems Inc., and

One additional U.S. employer that has chosen to remain anonymous in this initial report.

Members of the Community of Practice split into the following committees:

Current Practices, chaired by Alan Gardner of ING North America;

HR Systems, chaired by Chris Crespo of Ernst & Young LLP;

Data Utilization, chaired by Howard Solomon of Sun Microsystems Inc.;

Legal Issues, chaired by Terry Hildebrandt of Terry Hildebrandt and Associates LLC; and

Report Integration, chaired by Ralph Carter of Xerox Corp.

In addition to each of the above-named chairs, Pat Baillie of Out & Equal Workplace Advocates and Deena Fidas and Samir Luther of Human Rights Campaign Foundation contributed to the final wording of this report.

Additional plans for use of this report include developing best practices resources, quick reference lists, training materials, and an ongoing update as new companies begin to develop metrics to advance LGBT workplace equality.

The Community of Practice has grown to more than 80 professionals representing more than 35 U.S. and Canadian employers, and non-profit organizations including the Human Rights Campaign Foundation and Out & Equal Workplace Advocates. Get more information and join the Community online at: http://lgbtselfid.collectivex.com/