Question 680 A clinical analytics source records visit costs such as 86.75, 91.20, and -12.50 for refunded charges. The amount must support averages and sums, including cents. Which Power Query type fits without discarding fractional portions? Decimal Number or an appropriate fixed-decimal monetary type Whole Number to truncate cents at the source Date because the source column contains periods True/False based on whether each charge is positive Explanation Correct Answer A:A decimal-capable numeric type preserves fractional charges and refunds for arithmetic; fixed decimal may be appropriate when currency precision is required. Incorrect Answers B:Whole Number removes cents and can bias financial aggregates. C:Dates represent calendar values rather than prices. D:A Boolean sign flag loses the magnitude of charges and refunds.
Question 681 An HR semantic model contains a confidential Salary field. A model author hides Salary in the report field list and assumes that viewers with Build permission cannot retrieve it. What security fix is appropriate? Apply supported object-level security or exclude the sensitive column from the exposed model for unauthorized users Rely only on Hide in report view as an enforcement boundary Place a warning text box beside the hidden field Rename Salary to X1 and assume unauthorized DAX queries cannot access it Explanation Correct Answer A:Hiding a column is authoring usability, not an access-control mechanism. OLS or removing data prevents unauthorized retrieval through model queries. Incorrect Answers B:Build-capable users may still query model fields despite their hidden state. C:A warning does not prevent retrieval of protected values. D:Renaming obscures a label but leaves data accessible to query consumers.
Question 682 A retail operations report uses DirectQuery to a SQL Server data warehouse. Report interactions are delayed; SQL Server metrics show lock waits and heavy scans. What should the team prioritize before redesigning every Power BI visual? Examine generated SQL, source indexes and concurrency, then optimize the costly source queries Add report bookmarks for each employee to reduce SQL locks Change formatting themes to shrink server requests Enable continuous auto-page refresh so slow queries run more often Explanation Correct Answer A:DirectQuery sends work to the source. Query plans, indexes, aggregation and concurrent load are primary diagnostics when server scans and locks dominate. Incorrect Answers B:Bookmarks alter client report state and do not improve SQL query plans. C:Themes change presentation, not relational scans and lock waits. D:More frequent refresh can increase the pressure on an already overloaded source.
Question 683 A delivery table contains six HandlingMinutes values: 56, 59, 62, 55, 52, 49. The measure is COUNTROWS(FILTER(Fact, Fact[HandlingMinutes] <= 55)). What does it return? 6 because COUNTROWS always ignores FILTER 3 2 because 55 is excluded 4 because 56 is rounded down into range Explanation Correct Answer B:The values 55,52,49 meet the inclusive <=55 condition, yielding three rows. Incorrect Answers A:COUNTROWS evaluates the filtered table expression supplied to it. C:The comparison is inclusive, so 55 counts. D:56 exceeds 55 and cannot be rounded into a logical comparison without a specified transformation.
Question 684 A logistics report calculates FreightCharge as Quantity multiplied by PricePerUnit. The fact table has no stored FreightCharge column, and prices vary by row. Which DAX expression captures the total correctly? SUM(Fact[Quantity]) * SUM(Fact[PricePerUnit]) SUMX(Fact, Fact[Quantity] * Fact[PricePerUnit]) AVERAGE(Fact[Quantity]) with no price term FORMAT(SUM(Fact[Quantity]),"#,##0") Explanation Correct Answer B:The iterator multiplies values within each fact row before summing all line amounts. Incorrect Answers A:Multiplying independent column sums introduces cross-row combinations and overstates charges. C:Quantity alone omits price and cannot compute freight value. D:Formatting a quantity sum produces text, not line-level charge arithmetic.
Question 685 A certified semantic model now points to a new unreviewed data source. Its lineage and validation evidence have changed. What governance action should the certifier evaluate? Review the changed source and model evidence, and remove or withhold Certified status until the approval criteria are met Assume certification permanently guarantees any replacement source Turn on report animations so consumers do not notice the source change Rename the semantic model to Certified without an approval review Explanation Correct Answer A:Certification represents a controlled endorsement of reviewed content; a material source change requires renewed confidence in those criteria. Incorrect Answers B:A certification label is not proof that future unseen data sources have been checked. C:Animations cannot compensate for broken lineage or quality controls. D:Names and titles do not establish authorized endorsement.
Question 686 A food chain wants capable business users to change a chart from a region breakdown to a product breakdown in reading view, without editing or republishing the original report. The report author wants a curated selector offering Region, Product or Channel as alternative chart axes to all consumers, without requiring ad hoc personalized visual editing. What feature supports this design? Create a field parameter and connect it to the relevant visual and selector Give every consumer full report edit permissions Duplicate the report three times Create a separate Import dataset for every axis Explanation Correct Answer A:Field parameters provide controlled user-selectable fields within an authored visual. Incorrect Answers B:Editors rights are unnecessary for choosing from approved dimension options. C:Report duplication increases maintenance and fragments the experience. D:Reimporting data has no relationship to axis-field switching.
Question 687 A model contains CalendarDate and FiscalDate dimensions; both expose a column called Year. Self-service users keep applying the wrong Year slicer to accounting reports. What model-authoring change improves clarity? Rename/display the columns as Calendar Year and Fiscal Year and document their intended uses Delete the fiscal date table although financial reports use it Set all relationship filters to Both and assume users infer which Year they chose Change both Year fields to currency types Explanation Correct Answer A:Distinct field names communicate the time role without altering model relationships or removing needed date choices. Incorrect Answers B:Deleting the fiscal table removes the required accounting calendar. C:Filter-direction changes do not clarify field semantics. D:Currency typing introduces incorrect data meaning.
Question 688 A mobile-games analyst uses scatter x/y measures for spend and users, and employee count controls bubble area. One branch has 100 times the employee count of others and its bubble obscures most points. What is an appropriate report-design review? Check the bubble-size scale and whether the third measure adds interpretable information; adjust size encoding or choose an alternative view that exposes smaller branches Convert the x-axis amount to a Date to hide the outlier Delete all low-employment branches from the model so the large bubble remains Turn off source SQL authentication to shrink plotted symbols Explanation Correct Answer A:Bubble area can overwhelm other observations when one value dominates; reviewing the size field and scale protects interpretability without corrupting data. Incorrect Answers B:Changing numeric spending into a date type produces invalid semantics. C:Discarding legitimate smaller observations creates analytical bias. D:Authentication has no role in scatter marker sizing.
Question 689 A sensor fact contains one row per event with EventTimestamp and DeviceKey. A Device dimension also contains each device’s installation date. Which table should hold the event timestamp used for per-event trending? The event-grain fact table, with the Device dimension providing device attributes The Device dimension repeated once per event timestamp A report bookmark replacing source event time A new many-to-many Device key linking every timestamp as a device Explanation Correct Answer A:An event timestamp describes a fact occurrence; moving it into a one-row-per-device dimension would break the dimension grain. Incorrect Answers B:Repeating device dimension rows for events destroys its unique device key. C:A bookmark cannot replace a recorded timestamp. D:Turning timestamps into device members creates an invalid entity relationship.
Question 690 A charity’s branch scatter plot shows one branch with high donation counts but exceptionally low average donation value. Leadership wants to investigate that point without dropping it from reported totals. What should the analyst do? Replace the quantitative axes with two unrelated text IDs Set a constant reference line and assume it explains the unusual value Use a scatter plot with relevant axes and drillthrough or contextual tooltip to inspect the atypical branch Remove the point with a visual-level filter and publish the reduced total Explanation Correct Answer C:The outlying branch is an analytical finding. Its supporting rows should remain available so investigators can determine whether price mix or data quality explains it. Incorrect Answers A:Two text identifiers do not provide the numerical relationship being investigated. B:A fixed reference line can mark a threshold but cannot diagnose the branch’s transaction mix. D:Filtering the observation out hides a possible business problem and changes the evidence.
