对等比较查询Peer comparison queries

Workplace Analytics 对等比较查询可帮助您分析所选个人的工作区协作行为,具体方法是将其与一个或多个这些个人的对等组的协作行为进行比较。A Workplace Analytics peer comparison query helps you analyze the workplace collaboration behavior of selected individuals by comparing it with the collaboration behavior of one or more of those individuals' peer groups.

您可以使用对等比较查询来确定针对其他人以特定方式进行比较的人员。You use a peer comparison query to identify people who compare in particular ways against other people. 在生成查询时,可以确定感兴趣的个人、要与之进行比较的组、比较标准以及查询数据的一段时间。As you build the query, you identify the individuals of interest, the groups to compare them with, the comparison metrics, and a period of time for which to query data. 请注意,在查询各个用户时,在输出中不会标识它们;结果仅显示 PersonIds。Note that while you are querying about individuals, they are not identified in the output; results show only PersonIds.

虽然对等比较查询侧重于人员,但它们生成的信息与人员查询不同。Although peer comparison queries focus on people, they produce different information than person queries. 使用人员查询可了解个人的组织属性 — (如团队、级别或位置) — 以及他们使用时间的方式之间的关系,或者您希望了解其工作时间分配的一个方面可能会影响其时间分配的其他方面。Use a person query to understand the relationship between a person’s organizational attributes—such as their team, level, or location—and how they use their time, or when you want to know how one aspect of their on-the-job time allocation might influence other aspects of their time allocation. 使用对等比较查询,将此人的协作行为与此人的对等组中的行为进行比较。Use a peer comparison query to compare the person's collaboration behavior with that behavior in the person's peer groups.

创建对等比较查询Create a Peer comparison query

角色: 分析师Role: analyst

创建对等比较查询To create a peer comparison query

  1. 打开 工作区分析Open Workplace Analytics. 如果出现提示,请使用你的工作帐户登录。If prompted, sign in with your work account.

  2. 在左侧导航窗格中,展开 " 分析 ",然后选择 " 查询"。In the left navigation pane, expand Analyze and then select Queries.

  3. 在 " 查询 " 页上的 " 启动自定义查询" 下,选择 " 对等比较"。On the Queries page, under Start custom query, select Peer comparison. 这将打开 对等比较查询 页面:This opens the Peer Comparison query page:

    对等比较查询页

  4. 将查询命名并为其添加说明(可选)。Name the query and, optionally, add a description for it.

  5. 对于 " 分组依据",选择对 — — 数据进行聚合和报告所依据的一段时间(天、周或月)。For Group by, select an amount of time—day, week, or month—over which to aggregate and report on data.

  6. 对于 " 时间段",指定要查询其数据的时间段。For Time period, specify the time period whose data you want to query.

  7. (可选) 选择 " 自动刷新"。(Optional) Select Auto-refresh. (有关详细信息,请参阅 查询的自动刷新选项。 ) (For more information, see Auto-refresh option for queries.)

  8. 如果您的指标中至少有一个指标与会议有关,请考虑使用会议排除项来排除来自查询之外的不属于相关规范的会议。If at least one of your metrics has to do with meetings, consider using Meeting exclusions to exclude meetings that fall outside relevant norms from the queries. 您可以在默认会议排除规则之间进行选择,也可以创建与贵公司的会议约定相匹配的自定义规则。You can select between the default meeting exclusion rules or create custom rules that match your company’s meeting conventions. 若要了解详细信息,请参阅 会议排除See Meeting exclusions to learn more.

  9. 在 " 选择员工 " 部分,添加筛选器以选择作为查询主题的员工。In the Select employees section, add filters to select the employees who are the subject of your query. 有关详细信息,请参阅 选择员工For more information, see Select employees.

  10. 在 " 选择对等组 " 部分,选择您要与您的主题总体进行比较的员工 (您在前面步骤中选择的人) 。In the Select peer groups section, select the employees whom you want to compare with your subject population (the people you selected in the preceding step). 有关详细信息,请参阅 选择对等组For more information, see Select peer groups.

  11. 在 " 选择指标 " 部分中,选择要与对等组进行比较的员工行为的方面。In the Select metrics section, select the aspect of employee behavior that you want to compare with peer groups. 有关详细信息,请参阅 选择指标For more information, see Select metrics.

  12. 若要运行查询,请选择页面右上角的 " 运行 "。To run the query, select Run at the top right of the page. 将显示一条消息,指示查询正在运行。A message indicates that the query is running. 将显示 " 查询/结果 " 页,并指示所启动的查询的状态为 "正在运行":The Queries / Results page appears and indicates that the query you started has a status of "running":

    查询正在运行

概念Concepts

以下各节提供了可帮助您创建对等比较查询的信息。The following sections provide information that can help you as you create a peer comparison query.

选择员工Select employees

对等比较查询的目的是将特定员工的工作场所行为的各个方面与组织中的其他人进行比较,而将其与组织中的其他用户组进行比较。The purpose of a peer comparison query is to bring aspects of particular employees' workplace behavior into focus by comparing them with others in the organization, others in one of their peer groups. 在 " 选择员工 " 步骤中,选择这些人。In the Select employees step, you select these people. (还将 — — 在 " 选择对等组 " 步骤中选择要与之进行比较的其他人。 ) (You will also select the other people—the ones you're comparing them with—in the Select peer groups step.)

若要选择您要查询的人员,请按其 HR 属性对其进行筛选,如下所示:To select the people you’re querying about, filter them by their HR attributes, such as the following:

  • 域: 选择其电子邮件地址具有特定域的所有人员Domain: Select all the people whose email addresses have a particular domain
  • HourlyRate: 选择获得特定金额的所有员工HourlyRate: Select all employees who earn a particular amount
  • FunctionType: 选择,例如,所有的工程师或所有 HR 员工FunctionType: Select, for example, all engineers or all HR employees
  • 组织: 选择一个或多个组织(如设施和财务)中的所有人Organization: Select everyone in one or more organizations, such as Facilities and Finance

您可以以其他方式选择组织。You can select organizations in other ways. 例如,通过 — 指定芝加哥办事处( — 如果在组织数据中上载了地理属性)的地理位置。For example, by geography—by designating the Chicago office—if geographic attributes were uploaded in the organizational data.

即使您在 " 选择员工 " 步骤中执行的筛选可为您提供多个人员,您也不会选择一个组。Even though the filtering that you perform in the Select employees step provides you with multiple people, you are not selecting a group. 而是选择一个个人列表。Rather, you are selecting a list of individuals. 运行查询时,将为列表中的每个个体构造一个对等组,然后为列表中的每个单独项进行比较,最后为列表中的每个单独的结果显示结果。When the query runs, a peer group is constructed for each individual in the list, and then, a comparison is made for each individual in the list, and finally, results appear for each individual in the list.

您选择的多个用户与查询的运行方式无关,而不是向您提供有关这些人员的类似信息,报告在输出文件中的多个行上,每个聚合周期的每个人的一行,分别对应于每个聚合周期 (日、周或月) 其 HR 属性匹配您的筛选器参数。The fact that you have selected multiple individuals has no bearing on how the query runs, other than to provide you with similar information about those individuals, reported on multiple rows in the output file, one row for each person for each aggregation period (day, week, or month) whose HR attributes matched your filter parameters.

Note

由于您选择的是个人,而不是组,因此最小组大小设置不适用于您在此步骤中选择的员工列表。Because you have selected individuals, not groups, the minimum-group-size setting does not apply to the lists of employees that you select in this step. 如果此列表只包含一个雇员,则您仍有一个可以返回有效结果的有效查询。If this list contained only one employee, you'd still have a valid query that could return valid results.

选择对等组Select peer groups

在 " 选择员工 " 步骤中,选择您要查询的相关人员。In the Select employees step, you select individuals of interest, those you’re querying about. 在 " 选择对等组 " 步骤中,选择要与这些用户进行比较的组。In the Select peer groups step, you select the groups to compare with those individuals. 这些是一组类似的雇员,分别为每个选定的员工进行计算。These are groups of similar employees, calculated separately for each selected employee.

选择对等组

在此示例中,我们选择了用于指示要比较的组的选项。In this example, we’ve selected options to indicate groups to compare. 以下各节介绍了这些选项:The following sections describe those options:

与员工共用同一个经理Shares same manager as employee

在此示例中,我们在管理层次结构中选择了员工的直接同级,这是共享同一经理的人员。In this example, we selected the employee’s direct peers in the management hierarchy, the people who share the same manager. 对于每个感兴趣的人,其工作场所行为如何尤其是我们选择的指标所描述的行为的方位–与比较组的此方面进行了比较?For each individual of interest, how does their workplace behavior – in particular, the aspect of behavior described by the metric we’ve selected – compare with that aspect for the comparison group? 此比较结果将在每个聚合期间的一行中的查询结果文件中显示。The results of this comparison will appear in the query results file in one line for each aggregation period.

直接向员工报告Reports directly to employee

第二个复选框选项是 "直接向所选员工报告"。The second checkbox option is “Reports directly to the selected employee.” 此选项 (我们在此示例中未选择) 将感兴趣的个人的工作场所行为与向其报告的人员的工作场所行为进行比较。This option (which we did not select in this example) compares workplace behavior of the individual of interest to the workplace behavior of the people who report to them.

此比较的用途是什么?Of what use is this comparison? 例如,经理有效性的一个重要度量单位是经理的网络大小与员工的网络大小。For example, one important measure of manager effectiveness is the manager’s network size versus their employees’ network size. 通过将这些人员连接到组织中的其他人员,优质经理将增加其直接下属的网络大小。Good managers will increase the network size of their direct reports by connecting those people to other people in the organization. 选择此选项并指定 "网络大小" 指标,将经理的网络大小与他们的直接下属的网络大小进行比较。Select this option and specify a "network size" metric to compare the manager’s network size with that of their direct reports.

自定义比较组Custom comparison group

自定义比较组允许您使用组织属性定义您在组织中选择的组。The Custom comparison group lets you define a group of your choosing in the organization by using organizational attributes. 在此示例中,我们选择了在同一职业生涯中的员工组,并与感兴趣的个人具有相同的工作职务。In this example, we selected groups of employees who are at the same career stage and have the same work title as the individuals of interest.

例如,一个感兴趣的人是 "Emil"。For example, one individual of interest was "Emil." Emil 只是从学校进行了渐变,是一级工程师。Emil has just graduated from college and is a first-level engineer. 在此示例中, 自定义比较组 还由刚刚进行了分级的用户和第一级的工程师组成。In this example, the Custom comparison group also consists of people who’ve just graduated and are first-level engineers. 此比较结果将显示在查询结果文件中 Emil 的第二行中。The results of this comparison will appear in a second line for Emil in the query results file.

选择多个选项Selection of multiple options

由于我们在此示例中选择了两个组,因此当查询运行时将进行两项比较。Because we selected two groups in this example, two comparisons will be made when the query runs. 这意味着,对于每个感兴趣的人,每个聚合周期的 — 日、周或月的查询结果文件中都会出现两行结果。This means that, for each individual of interest, two lines of results will appear in the query results file for each aggregation period—day, week, or month.

Note

由于在 " 选择对等组 " 步骤中,选择了 "组,而不是个人",因此应用了最小组大小设置。Because in the Select peer groups step you select groups, not individuals, the minimum-group-size setting does apply. 在此步骤中选择的任何组都必须满足或超过对贵组织有效的最小组大小。Any groups you select in this step must meet or exceed the minimum group size that is in effect for your organization.

选择指标Select metrics

在 " 选择指标 " 步骤中,首先选择一个或多个指标。In the Select metrics step, you start by selecting one or more metrics. 例如,谁有更大的 "内部网络大小" (指标) 、员工或比较组?For example, who has the larger "internal network size" (a metric), the employees or the comparison groups? 在下图中,我们选择了四个指标,这些指标在 " 基本指标" 中列出:In the following illustration, we’ve selected four metrics, which you see listed under Base metric:

选择指标

选择基本指标后,可以更改其显示名称。After you select a base metric, you can change its display name.

选择 "指标选项"Select metrics options

在本节中,您将定义在比较中使用指标的方式:In this section, you define how the metrics are used in the comparison:

  • 对于 " 计算类型",选择是比较均值值、中间值还是另一个统计运算符。For Calculation type, select whether to compare the mean value, the median, or another statistical operator.
  • 在 " 差异类型" 中,选择是以绝对值还是以百分比表示此比较的结果。For Difference type, select whether to express the result of this comparison in absolute numbers or in percentages.

您可以在所选的指标中改变计算类型和差异类型。You can vary the Calculation types and Difference types in the metrics that you’ve chosen.

查看结果View results

查询完成运行后,其 " 查询/结果 " 页面中的行会在 "状态" 列中显示一个复选标记。After the query finishes running, its row in the Queries / Results page shows a check mark in the Status column. 然后,您可以下载并打开结果文件。You can then download and open the results file. (结果文件采用 .csv 格式。 ) (Results files are in .csv format.)

结果文件

在此示例结果文件中,第12行数据 (号为2到13的行) 报告第一个利息的5A329 的比较结果,其 PersonId 结束于 ""。In this example results file, the first 12 lines of data (lines numbered 2 through 13) report comparison results for the first individual of interest, whose PersonId ends in "5A329."

结果文件详细信息

下面是我们了解的有关此比较的一些信息, (这12行) 上的结果:Here are a few things we know about this comparison (the results on these 12 lines):

  • PeerGroup值告诉我们此人与共享同一个经理的同事进行了比较。The PeerGroup value tells us that this person was compared with peers who share the same manager.
  • Date值指出从中提取数据的聚合期间的第一天。The Date value states the first day of the aggregation period from which data was drawn.
  • 在这种情况下,这12行中的每一行代表一个聚合期间 — (一周),因为我们在定义查询时在 Group by 控件中选择了 "周"。Each of these 12 lines represents one aggregation period—in this case, one week, because we selected "week" in the Group by control while defining the query.

在从1月27日开始的一周内,用户的对等组平均发送了126的用户发送的140电子邮件。The person sent 140 emails during the week that started in January 27, while their peer group sent 126, on average. 区别在于14。The difference was 14. 但是,在下一周的 (从2月3日开始) ,发送109的人和他们的组平均120,因此 Emails_sent_Difference 表明他们发送了11封电子邮件,而不超过其对等组。However, during the following week (which began on February 3), the person sent 109 and their group sent on average 120, so Emails_sent_Difference shows that they sent 11 fewer emails than their peer group. 如果您选择了 " 差异类型的百分比",则显示的结果将为 "-9%",而不是由其对等组发送的) 电子邮件的数量减少了 (九%。If you had chosen percentage for Difference type, the displayed result would have been "-9%" (nine percent fewer) emails than were sent by their peer group.

有关详细信息,请参阅 查看、下载和导出查询结果For more information, see View, download, and export query results.

了解并解释查询输出Understand and interpret query output

指标说明Metric descriptions

查看、下载和导出查询结果View, download, and export query results